hash stringlengths 32 32 | doc_id stringlengths 5 12 | section stringlengths 5 1.47k | content stringlengths 0 6.67M |
|---|---|---|---|
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.4.14 Support of UE-Satellite-UE communication | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.4.14.1 General | To support UE-Satellite-UE communication, network functions (e.g. UPF) transferring IMS user plane traffic in the 5G system shall be deployed onboard the satellites with regenerative payload that provide service link to the two UEs. The reference architecture for support of UE-Satellite-UE communication is described in TS 23.228 [15].
NOTE 1: In this Release, only IMS voice/video services between two UEs belonging to same PLMN and in non-roaming scenario are considered.
To keep the UE's IP address for IMS services unchanged, a UPF on the ground shall be selected during PDU Session establishment procedure as the PSA UPF for the duration of the IMS PDU session. The UPF on-board the satellite shall only be selected as an UL CL/BP and local PSA to route IMS user plane traffic.
NOTE 2: In this Release of the specification, the type of serving satellite can only be LEO or MEO. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.4.14.2 Setup of UE-Satellite-UE communication with IMS user plane components on-board satellites | The AMF reports the identifier of the satellite serving the UE to the SMF as part of the PDU Session establishment and update procedures as described in clauses 4.3.2 and 4.3.3 of TS 23.502 [3].
NOTE 1: The AMF can determine that the UE is accessing gNB onboard a LEO or MEO satellite and can also determine the satellite identifier based on local configuration, e.g. Global RAN Node IDs associated with the LEO or MEO satellite.
Upon receiving a request from the P-CSCF for accessing network information report (e.g. as a result of the P-CSCF processing an IMS call setup), the PCF subscribes to Access Network Information report from the SMF for the IMS PDU Session, as described in clause 6.1.3 of TS 23.503 [45]. The Access Network Information report contains the identifier of the satellite serving the UE as reported by the AMF.
NOTE 2: How the serving satellite identifier is used by the IMS to determine whether the UE-Satellite-UE communication can be activated is specified in TS 23.228 [15]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.4.14.3 Support for UE-satellite-UE communication when serving satellite changes | NOTE 1: The following mechanism assumes gNB, UL CL/L-PSA, IMS-AGW are onboard the same satellite.
In the UE-SAT-UE communication, if the UE's serving satellite changes (e.g. due to UE movement or the satellite movement), the Xn/N2-based handover procedure can be performed to change the serving gNB from that on the source satellite to that on the target satellite. The AMF includes the target satellite ID in the Nsmf_PDUSession_UpdateSMContext Request to the SMF. The SMF decides whether to route IMS user plane traffic to the PSA on the ground or retain the UL CL/L-PSA on the satellite based on local configuration and other factors as specified in clause 6.3.3.
NOTE 2: The SMF can decide to route the IMS user plane traffic to the PSA on ground, e.g., due to no connection between the source satellite and the target satellite.
If the SMF decides to route the IMS user plane traffic to the PSA on the ground and the procedure for removing the UL CL/L-PSA as described in clause 4.3.5.5 of TS 23.502 [3] is to be performed, the SMF provides an early notification without satellite ID via Nsmf_EventExposure_Notify to PCF. The PCF forwards this notification to the P-CSCF via Npcf_PolicyAuthorisation_Notify, indicating that the 5GC user plane needs to fall back to the ground.
If the SMF decides to retain the UL CL/L-PSA on the satellite, it provides an early notification with the target satellite ID via Nsmf_EventExposure_Notify to PCF and the PCF forwards this target satellite ID to the P-CSCF via Npcf_PolicyAuthorisation_Notify. In this case, the IMS informs 5GC whether the UE-SAT-UE communication can or cannot continue, as outlined in clause AE.5.2.1 of TS 23.228 [15], i.e. the PCF receives a response from the P-CSCF and forwards it to the SMF:
- If the response indicates that the UE-SAT-UE communication can continue, the SMF allocates the UL CL/BP and L-PSA on the target satellite by performing a simultaneous change of UL CL/BP and additional PSA for the PDU Session, as described in clause 4.3.5.7 of TS 23.502 [3], where the UL CL/BP and L-PSA rule shall be updated with the IMS-derived transport address and optional N6 traffic routing information associated with target DNAI as received in the response. A late notification is sent by SMF to the P-CSCF via PCF afterwards, as outlined in clause 4.3.6.3 of TS 23.502 [3].
- If the response indicates that the UE-SAT-UE communication cannot continue, the SMF removes the UL CL/BP and L-PSA on the source satellite by following the procedure described in clause 4.3.5.5 of TS 23.502 [3]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5 Non-3GPP access specific aspects | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.0 General | This clause describe the specific aspects for untrusted non-3GPP access, trusted non-3GPP access and W-5GAN access. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.1 Registration Management | This clause applies to Non-3GPP access network corresponding to the Untrusted Non-3GPP access network, to the Trusted Non-3GPP access network and to the W-5GAN. In the case of W-5GAN the UE mentioned in this clause corresponds to 5G-RG or to the W-AGF in the case of FN-RG. In the case of N5CW devices access 5GC via trusted WLAN access networks, the UE mentioned in this clause corresponds to TWIF.
The UE shall enter RM-DEREGISTERED state and the AMF shall enter RM-DEREGISTERED state for the UE on non-3GPP access as follows:
- at the UE and at the AMF, after performing an Explicit Deregistration procedure;
- at the AMF, after the Network non-3GPP Implicit Deregistration timer has expired.
- at the UE, after the UE non-3GPP Deregistration timer has expired.
NOTE: This is assumed to leave sufficient time to allow the UE to re-activate UP connections for the established PDU Sessions over 3GPP or non-3GPP access.
Whenever a UE registered over non-3GPP access enters CM-IDLE state for the non-3GPP access, it starts the UE non-3GPP Deregistration timer according to the value received from the AMF during a Registration procedure.
Over non-3GPP access, the AMF runs the Network non-3GPP Implicit Deregistration timer. The Network non-3GPP Implicit Deregistration timer is started with a value longer than the UE's non-3GPP Deregistration timer, whenever the CM state for the UE registered over non-3GPP access changes to CM-IDLE for the non-3GPP access.
For a UE that is registered over Non-3GPP access, a change of the point of attachment (e.g. change of WLAN AP) shall not lead the UE to perform a Registration procedure.
A UE that is registered over non-3GPP access may trigger a Mobility Registration Update procedure via a new non-3GPP AN node (i.e. N3IWF or TNGF) to switch traffic from an old non-3GPP access to a new non-3GPP access.
Traffic switching from an old non-3GPP access to and from a wireline access is not supported in this Release.
Traffic switching from an old non-3GPP access to new non-3GPP access is supported only when PLMN of the new non-3GPP access is the same PLMN of the old non-3GPP access.
A UE shall not provide 3GPP-specific parameters (e.g. indicate a preference for MICO mode) during registration over a non-3GPP access.
During registration procedure the AMF may determine whether the serving N3IWF/TNGF is appropriate based on the slices supported by the N3IWFs/TNGFs as specified in clause 6.3.6 and clause 6.3.12 respectively. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.2 Connection Management | This clause applies to Non-3GPP access network corresponding to the Untrusted Non-3GPP access network, to the Trusted Non-3GPP access network and to the W-5GAN. The UE mentioned in this clause corresponds to the 5G-RG in the case of W-5GAN and to the W-AGF in the case of FN-RG. In the case of N5CW devices access 5GC via trusted WLAN access networks, the UE mentioned in this clause corresponds to TWIF.
A UE that successfully establishes a Non-3GPP Access Connection to the 5GC over a Non-3GPP access transitions to CM-CONNECTED state for the Non-3GPP access.
In the case of Untrusted Non-3GPP access to 5GC, the Non-3GPP Access Connection corresponds to an NWu connection.
In the case of Trusted access to 5GC, the Non-3GPP Access Connection corresponds to an NWt connection.
In the case of N5CW devices access 5GC via trusted WLAN access networks, the Non-3GPP Access Connection corresponds to an Yt' connection.
In the case of Wireline access to 5GC, the Non-3GPP Access Connection corresponds to a Y4 connection and to Y5 connection.
A UE does not establish multiple simultaneous Non-3GPP Access Connection to the 5GC.
The Non-3GPP Access Connection is released either as a result of an Explicit Deregistration procedure or an AN Release procedure.
In the case of Untrusted Non-3GPP access, Trusted Non-3GPP access and W-5GAN access to 5GC, the N3IWF, TNGF, TWIF and W-AGF may in addition explicitly release the NWu, NWt, Yt', Y4 and Y5 signalling connection due to NWu, NWt, Yt', Y4 and Y5 connection failure, respectively. In the case of NWu and NWt, the release may be determined by the "dead peer detection" mechanism in IKEv2 defined in RFC 7296 [60]. In the case of Y4 and Y5 the release may be detected for example by lost of synchronisation of physical link, lost of PPPoE session, etc. Further details on how NWu, NWt, Yt', Y4 and Y5 connection failure is detected is out of scope of 3GPP specifications.
For W-5GCAN, the W-AGF explicitly releases the N2 connection due to Y4 or Y5 connection failure, as determined by the "dead peer detection" mechanism in DOCSIS MULPI [89].
The release of the Non-3GPP Access Connection between the UE and the N3IWF, TNGF, TWIF or W-AGF shall be interpreted as follows:
- By the N3IWF, TNGF, TWIF and W-AGF as a criterion to release the N2 connection.
- By the UE as a criterion for the UE to transition to CM-IDLE. A UE registered over non-3GPP access remains in RM-REGISTERED state, unless the Non-3GPP Access Connection release occurs as part of a Deregistration procedure over non-3GPP access in which case the UE enters the RM-DEREGISTERED state. When the UE in RM-REGISTERED transitions to CM-IDLE, the UE non-3GPP Deregistration timer starts running in the UE. The UE non-3GPP Deregistration timer stops when the UE moves to CM-CONNECTED state or to the RM-DEREGISTERED state.
NOTE 1: When moved to CM-IDLE state over one access, the UE can attempt to re-activate UP connections for the PDU Sessions over other access, per UE policies and depending on the availability of these accesses.
NOTE 2: The release of the NWu, NWt, Yt', Y4 or Y5 at the UE can occur as a result of explicit signalling from the N3IWF, TNGF, TWIF or W-AGF respectively, e.g. IKE INFORMATION EXCHANGE in the case of NWu or as a result of the UE detecting NWu, NWt, Yt', Y4 or Y5 connection failure, e.g. as determined by the "dead peer detection" mechanism in IKEv2 as defined in RFC 7296 [60] for NWu, NWt and Yt' or W-5GAN access specific mechanism for Y4 and Y5. Further details on how the UE detects NWu, NWt, Yt', Y4 or Y5 connection failure is out of scope of 3GPP specifications.
In the case of Non-3GPP access, when the AMF releases the N2 interface, the N3IWF, TNGF, TWIF and W-AGF shall release all the resources associated with the UE including the Non-3GPP Access Connection with the UE and its corresponding N3 resources. A release of the N2 connection by the AMF shall set the CM state for the UE in the AMF to CM-IDLE.
NOTE 3: It is assumed that a UE configured to receive services from a 5GC over non-3GPP access that is RM-DEREGISTERED or CM-IDLE over the non-3GPP access will attempt to establish Non-3GPP Access Connection and transition to CM-CONNECTED state whenever the UE successfully connects to a non-3GPP access unless prohibited by the network to make a N3GPP Access Connection (e.g. due to network congestion).
An UE cannot be paged on Non-3GPP access network.
When a UE registered simultaneously over a 3GPP access and a non-3GPP access moves all the PDU Sessions to one of the accesses, whether the UE initiates a Deregistration procedure in the access that has no PDU Sessions is up to the UE implementation.
Release of PDU Sessions over the non-3GPP access does not imply the release of N2 connection.
When the UE has PDU Sessions routed over the non-3GPP access and the UE state becomes CM-IDLE for the non-3GPP access, these PDU Sessions are not released to enable the UE to move the PDU Sessions over the 3GPP access based on UE policies. The core network maintains the PDU Sessions but deactivates the N3 user plane connection for such PDU Sessions. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.3 UE Reachability | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.3.1 UE reachability in CM-IDLE | This clause applies to Non-3GPP access network corresponding to the Untrusted Non-3GPP access network, to the Trusted Non-3GPP access network and to the W-5GAN. The UE mentioned in this clause corresponds to 5G-RG, in the case of W-5GAN or to W-AGF in the case of support of FN-RG. In the case of N5CW devices access 5GC via trusted WLAN access networks, the UE mentioned in this clause corresponds to TWIF.
An UE cannot be paged over Non-3GPP access network.
If the UE states in the AMF are CM-IDLE and RM-REGISTERED for the non-3GPP access, there may be PDU Sessions that were last routed over the non-3GPP access and without user plane resources. If the AMF receives a message with a Non-3GPP Access Type indication from an SMF for a PDU Session corresponding to a UE that is CM-IDLE for non-3GPP access and the UE is registered over 3GPP access in the same PLMN as the one registered over non-3GPP access, a Network Triggered Service Request may be performed over the 3GPP access independently of whether the UE is CM-IDLE or CM-CONNECTED over the 3GPP access. In this case, the AMF provides an indication that the procedure is related to non-3GPP access, as specified in clause 5.6.8.
NOTE: The UE behaviour upon such network triggered Service Request is specified in clause 5.6.8. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.5.3.2 UE reachability in CM-CONNECTED | This clause applies to Non-3GPP access network corresponding to the Untrusted Non-3GPP access network, to the Trusted Non-3GPP access network and to the W-5GAN. In the case of W-5GAN the UE mentioned in this clause corresponds to 5G-RG and to W-AGF in the case of support of FN-RG. In the case of N5CW devices access 5GC via trusted WLAN access networks, the UE mentioned in this clause corresponds to TWIF.
For a UE in CM-CONNECTED state:
- the AMF knows the UE location on a N3IWF, TNGF, TWIF and W-AGF node granularity.
- the N3IWF, TNGF, TWIF and W-AGF releases the N2 connection when UE becomes unreachable from N3IWF, TNGF, TWIF and W-AGF point of view, i.e. upon Non-3GPP Access Connection release. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6 Session Management | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.1 Overview | The 5GC supports a PDU Connectivity Service i.e. a service that provides exchange of PDUs between a UE and a data network identified by a DNN. The PDU Connectivity Service is supported via PDU Sessions that are established upon request from the UE.
The Subscription Information for each S-NSSAI may contain a Subscribed DNN list and one default DNN. When the UE does not provide a DNN in a NAS Message containing PDU Session Establishment Request for a given S-NSSAI, the serving AMF determines the DNN for the requested PDU Session by selecting the default DNN for this S-NSSAI if a default DNN is present in the UE's Subscription Information; otherwise the serving AMF selects a locally configured DNN for this S-NSSAI.
The expectation is that the URSP in the UE is always up to date using the procedure defined in clause 4.16.12.2 of TS 23.502 [3] and therefore the UE requested DNN will be up to date.
In order to cover cases that UE operates using local configuration, but also other cases where operator policies can be used in order to replace an "up to date" UE requested DNN with another DNN used only internally in the network, during UE Registration procedure the PCF may indicate, to the AMF, the operator policies to be used at PDU Session Establishment for DNN replacement of a UE requested DNN. PCF may indicate a policy for DNN replacement of UE requested DNNs not supported by the network and/or indicate a list of UE requested DNNs per S-NSSAI valid for the serving network, that are subject for replacement (details are described in TS 23.503 [45]).
If the DNN provided by the UE is not supported by the network and AMF cannot select an SMF by querying NRF, the AMF shall reject the NAS Message containing PDU Session Establishment Request from the UE with a cause indicating that the DNN is not supported unless the PCF provided the policy to perform a DNN replacement of unsupported DNNs.
If the DNN requested by the UE is indicated for replacement or the DNN provided by the UE is not supported by the network and the PCF provided the policy to perform DNN replacement of UE requested DNNs not supported by the network, the AMF shall interact with the PCF to perform a DNN replacement. During PDU Session Establishment procedure and as a result of DNN replacement, the PCF provides the selected DNN that is applicable for the S-NSSAI requested by the UE at the PDU Session Establishment. The AMF uses the selected DNN in the query towards the NRF for the SMF selection, as specified in clause 6.3.2 and provides both requested and selected DNN to the selected SMF. For PDU Session with Home-routed Roaming whether to perform DNN replacement is based on operator agreements.
NOTE 1: The selected DNN is determined based on operator preferences and can differ from subscribed DNNs. The matching of selected DNN to S-NSSAI is assumed to be based on network configuration.
Each PDU Session supports a single PDU Session type i.e. supports the exchange of a single type of PDU requested by the UE at the establishment of the PDU Session. The following PDU Session types are defined: IPv4, IPv6, IPv4v6, Ethernet, Unstructured.
PDU Sessions are established (upon UE request), modified (upon UE and 5GC request) and released (upon UE and 5GC request) using NAS SM signalling exchanged over N1 between the UE and the SMF. Upon request from an Application Server, the 5GC is able to trigger a specific application in the UE. When receiving that trigger message, the UE shall pass it to the identified application in the UE. The identified application in the UE may establish a PDU Session to a specific DNN, see clause 4.4.5.
SMF may support PDU Sessions for LADN where the access to a DN is only available in a specific LADN service area. This is further defined in clause 5.6.5.
SMF may support PDU Sessions for a 5G VN group which offers a virtual data network capable of supporting 5G LAN-type service over the 5G system. This is further defined in clause 5.8.2.13.
The SMF is responsible of checking whether the UE requests are compliant with the user subscription. For this purpose, it retrieves and requests to receive update notifications on SMF level subscription data from the UDM. Such data may indicate per DNN and per S-NSSAI of the HPLMN:
- The allowed PDU Session Types and the default PDU Session Type.
- The allowed SSC modes and the default SSC mode.
- QoS Information (refer to clause 5.7): the subscribed Session-AMBR, Default 5QI and Default ARP.
- The IP Index information.
- The static IP address/prefix.
- The subscribed User Plane Security Policy.
- the Charging Characteristics to be associated with the PDU Session. Whether this information is provided by the UDM to a SMF in another PLMN (for PDU Sessions in LBO mode) is defined by operator policies in the UDM/UDR.
NOTE 2: The content of the Charging Characteristics as well as the usage of the Charging Characteristics by the SMF are defined in TS 32.255 [68].
A PDU Session may support:
(a) a single-access PDU Connectivity Service, in which case the PDU Session is associated with a single access type at a given time, i.e. either 3GPP access or non-3GPP access; or
(b) a multi-access PDU Connectivity Service, in which case the PDU Session is simultaneously associated with both 3GPP access and non-3GPP access and simultaneously associated with two independent N3/N9 tunnels between the PSA and RAN/AN.
A PDU Session supporting a single-access PDU Connectivity Service is also referred to as single-access PDU Session, while a PDU Session supporting a multi-access PDU Connectivity Service is referred to as Multi-Access PDU (MA PDU) Session and it is used to support the ATSSS feature (see clause 5.32 for details).
A UE that is registered over multiple accesses chooses over which access to establish a PDU Session. As defined in TS 23.503 [45], the HPLMN may send policies to the UE to guide the UE selection of the access over which to establish a PDU Session.
NOTE 3: In this Release of the specification, at any given time, a PDU Session is routed over only a single access network, unless it is an MA PDU Session in which case it can be routed over one 3GPP access network and one Non 3GPP access network concurrently.
A UE may request to move a single-access PDU Session between 3GPP and Non 3GPP accesses. The decision to move single-access PDU Sessions between 3GPP access and Non 3GPP access is made on a per PDU Session basis, i.e. the UE may, at a given time, have some PDU Sessions using 3GPP access while other PDU Sessions are using Non 3GPP access.
If the UE is attempting to move a single-access PDU session from 3GPP access to non-3GPP access and the PDU session is associated with control plane only indication, then the AMF shall reject the PDU Session Establishment request as related CIoT 5GS optimisation features are not supported over non-3GPP access as described in clause 5.4.5.2.5 of TS 24.501 [47]. If the UE is attempting to move a single-access PDU session from non-3GPP access to NB-N1 mode of 3GPP access, the PDU Session Establishment request would also be rejected by AMF when the UP resources for the UE exceed the maximum number of supported UP resources as described in clause 5.4.5.2.4 of TS 24.501 [47].
In a PDU Session Establishment Request message sent to the network, the UE shall provide a PDU Session ID. The PDU Session ID is unique per UE and is the identifier used to uniquely identify one of a UE's PDU Sessions. The PDU Session ID shall be stored in the UDM to support handover between 3GPP and non-3GPP access when different PLMNs are used for the two accesses. The UE also provides as described in TS 24.501 [47]:
(a) PDU Session Type.
(b) S-NSSAI of the HPLMN that matches the application (that is triggering the PDU Session Request) within the NSSP in the URSP rules or within the UE Local Configuration as defined in clause 6.1.2.2.1 of TS 23.503 [45].
NOTE 4: If the UE cannot determine any S-NSSAI after performing the association of the application to a PDU Session, then it does not indicate any S-NSSAI in the PDU Session Establishment procedure as defined in clause 5.15.5.3.
(c) S-NSSAI of the Serving PLMN from the Allowed NSSAI, corresponding to the S-NSSAI of the HPLMN (b).
NOTE 5: In non-roaming scenario the mapping of the Allowed NSSAI to HPLMN S-NSSAIs is not provided to the UE (because the S-NSSAI of the Serving PLMN (c) has the same value of the S-NSSAI of the HPLMN (b)), therefore the UE provides in the PDU Session Request only the S-NSSAI of the Serving PLMN (c).
NOTE 6: In roaming scenarios the UE provides in the PDU Session Request both the S-NSSAI of the HPLMN (b) and the S-NSSAI of the VPLMN from the Allowed NSSAI (c) that maps to the S-NSSAI of the HPLMN.
(d) DNN (Data Network Name).
(e) SSC mode (Service and Session Continuity mode defined in clause 5.6.9.2).
Additionally, if the UE supports ATSSS and wants to activate a MA PDU Session, the UE shall provide Request Type as "MA PDU Request" and shall indicate the supported ATSSS capabilities (see clause 5.32 for details).
Table 5.6.1-1: Attributes of a PDU Session
PDU Session attribute
May be modified later during the lifetime of the PDU Session
Notes
S-NSSAI of the HPLMN
No
(Note 1) (Note 2)
S-NSSAI of the Serving PLMN
Yes
(Note 1) (Note 2) (Note 4)
DNN (Data Network Name)
No
(Note 1) (Note 2)
PDU Session Type
No
(Note 1)
SSC mode
No
(Note 2)
The semantics of Service and Session Continuity mode is defined in clause 5.6.9.2
PDU Session Id
No
User Plane Security Enforcement information
No
(Note 3)
Multi-access PDU Connectivity Service
No
Indicates if the PDU Session provides multi-access PDU Connectivity Service or not.
NOTE 1: If it is not provided by the UE, the network determines the parameter based on default information received in user subscription. Subscription to different DNN(s) and S-NSSAI(s) may correspond to different default SSC modes and different default PDU Session Types
NOTE 2: S-NSSAI(s) and DNN are used by AMF to select the SMF(s) to handle a new session. Refer to clause 6.3.2. The DNN may include both the Network Identifier and the Operator Identifier, see TS 29.502 [36]. See more details of the DNN usage and applicability, e.g. when full DNN or only Network Identifier is applied, in relevant stage 3 specifications.
NOTE 3: User Plane Security Enforcement information is defined in clause 5.10.3.
NOTE 4: The S-NSSAI value of the Serving PLMN associated to a PDU Session can change whenever the UE moves to a different PLMN, while keeping that PDU Session.
Subscription Information may include a wildcard DNN per subscribed S-NSSAI: when a wildcard DNN is associated with a subscribed S-NSSAI, the subscription allows, for this S-NSSAI, the UE to establish a PDU Session using any DNN value.
NOTE 7: The SMF is made aware whether the DNN of a PDU Session being established corresponds to an explicitly subscribed DNN or corresponds to a wildcard DNN. Thus, the SMF can reject a PDU Session establishment if the DNN of the PDU Session is not part of explicitly subscribed DNN(s) and local policies in the SMF require UE to have a subscription to this DNN.
A UE may establish multiple PDU Sessions, to the same data network or to different data networks, via 3GPP and via and Non-3GPP access networks at the same time.
A UE may establish multiple PDU Sessions to the same Data Network and served by different UPF terminating N6.
A UE with multiple established PDU Sessions may be served by different SMF.
The SMF shall be registered and deregistered on a per PDU Session granularity in the UDM.
The user plane paths of different PDU Sessions (to the same or to different DNN) belonging to the same UE may be completely disjoint between the AN and the UPF interfacing with the DN.
When the SMF cannot control the UPF terminating the N3 interface used by a PDU Session and SSC mode 2/3 procedures are not applied to the PDU Session, an I-SMF is inserted between the SMF and the AMF and handling of PDU Session(s) is described in clause 5.34.
NOTE 8: User Plane resources for PDU Sessions of a UE, except for regulatory prioritized service like Emergency Services and MPS, can be deactivated by the SMF if the UE is only reachable for regulatory prioritized services.
The SMF serving a PDU session (i.e. Anchor) can be changed during lifetime of the PDU session either within the same SMF set or, if the Context Transfer Procedures as specified in clause 4.26 of TS 23.502 [3] are supported, between SMFs in different SMF sets. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.2 Interaction between AMF and SMF | The AMF and SMF are separate Network Functions.
N1 related interaction with SMF is as follows:
- The single N1 termination point is located in AMF. The AMF forwards SM related NAS information to the SMF based on the PDU Session ID in the NAS message. Further SM NAS exchanges (e.g. SM NAS message responses) for N1 NAS signalling received by the AMF over an access (e.g. 3GPP access or non-3GPP access) are transported over the same access.
- The serving PLMN ensures that subsequent SM NAS exchanges (e.g. SM NAS message responses) for N1 NAS signalling received by the AMF over an access (e.g. 3GPP access or non-3GPP access) are transported over the same access.
- SMF handles the Session management part of NAS signalling exchanged with the UE.
- The UE shall only initiate PDU Session Establishment in RM-REGISTERED state.
- When a SMF has been selected to serve a specific PDU Session, AMF has to ensure that all NAS signalling related with this PDU Session is handled by the same SMF instance.
- Upon successful PDU Session Establishment, the AMF and SMF stores the Access Type that the PDU Session is associated.
N11 related interaction with SMF is as follows:
- The AMF reports the reachability of the UE based on a subscription from the SMF, including:
- The UE location information with respect to the area of interest indicated by the SMF.
- The SMF indicates to AMF when a PDU Session has been released.
- Upon successful PDU Session Establishment, AMF stores the identification of serving SMF of UE and SMF stores the identification of serving AMF of UE including the AMF set. When trying to reach the AMF serving the UE, the SMF may need to apply the behaviour described for "the other CP NFs" in clause 5.21.
N2 related interaction with SMF is as follows:
- Some N2 signalling (such as handover related signalling) may require the action of both AMF and SMF. In such case, the AMF is responsible to ensure the coordination between AMF and SMF. The AMF may forward the SM N2 signalling towards the corresponding SMF based on the PDU Session ID in N2 signalling.
- SMF shall provide PDU Session Type together with PDU Session ID to NG-RAN, in order to facilitate NG-RAN to apply suitable header compression mechanism to packet of different PDU type. Details refer to TS 38.413 [34].
N3 related interaction with SMF is as follows:
- Selective activation and deactivation of UP connection of existing PDU Session is defined in clause 5.6.8.
N4 related interaction with SMF is as follows:
- When it is made aware by the UPF that some DL data has arrived for a UE without downlink N3 tunnel information, the SMF interacts with the AMF to initiate Network Triggered Service Request procedure. In this case, if the SMF is aware that the UE is unreachable or if the UE is reachable only for regulatory prioritized service and the PDU Session is not for regulatory prioritized service, then the SMF shall not inform DL data notification to the AMF
The AMF is responsible of selecting the SMF per procedures described in clause 6.3.2. For this purpose, it gets subscription data from the UDM that are defined in that clause. Furthermore, it retrieves the subscribed UE-AMBR from the UDM and optionally dynamic serving network UE-AMBR from PCF based on operator local policy and sends to the (R)AN as defined in clause 5.7.2
AMF-SMF interactions to support LADN or LADN per DNN and S-NSSAI are defined in clause 5.6.5 and in clause 5.6.5a.
In order to support charging and to fulfil regulatory requirement (in order to provide NPLI (Network Provided Location Information) as defined in TS 23.228 [15]) related with the set-up, modification and release of IMS Voice calls or with SMS transfer the following applies
- At the time of the PDU Session Establishment, the AMF provides the SMF with the PEI of the UE if the PEI is available at the AMF.
- When it forwards UL NAS or N2 signalling to a peer NF (e.g. to SMF or to SMSF) or during the UP connection activation of a PDU Session, the AMF provides any User Location Information it has received from the 5G-AN as well as the Access Type (3GPP - Non 3GPP) of the AN over which it has received the UL NAS or N2 signalling. The AMF also provides the corresponding UE Time Zone. In addition, in order to fulfil regulatory requirement (i.e. providing Network Provided Location Information (NPLI), as defined in TS 23.228 [15]) when the access is non-3GPP, the AMF may also provide the last known 3GPP access User Location Information with its age, if the UE is still attached to the same AMF for 3GPP access (i.e. valid User Location Information).
The User Location Information, the access type and the UE Time Zone may be further provided by SMF to PCF. The PCF may get this information from the SMF in order to provide NPLI to applications (such as IMS) that have requested it.
The User Location Information may correspond to:
- In the case of 3GPP access: TAI, Cell-Id. The AMF includes only the Primary Cell-Id even if it had received also the Cell-Id of the Primary cell in the Secondary RAN node from NG-RAN.
- In the case of Untrusted non-3GPP access: TAI, the UE local IP address used to reach the N3IWF and optionally the UDP source port number if NAT is detected.
- In the case of Trusted non-3GPP access: TAI, TNAP/TWAP Identifier, the UE/N5CW device local IP address used to reach the TNGF/TWIF and optionally the UDP source port number if NAT is detected.
When the UE uses WLAN based on IEEE 802.11 technology to reach the TNGF, the TNAP Identifier shall include the SSID of the access point to which the UE is attached. The TNAP Identifier shall include at least one of the following elements, unless otherwise determined by the TWAN operator's policies:
- the BSSID (see IEEE Std 802.11-2012 [106]);
- civic address information of the TNAP to which the UE is attached.
The TWAP Identifier shall include the SSID of the access point to which the NC5W is attached. The TWAP Identifier shall also include at least one of the following elements, unless otherwise determined by the TWAN operator's policies:
- the BSSID (see IEEE Std 802.11-2012 [106]);
- civic address information of the TWAP to which the UE is attached.
NOTE 1: The SSID can be the same for several TNAPs/TWAPs and SSID only cannot provide a location, but it might be sufficient for charging.
NOTE 2: the BSSID associated with a TNAP/TWAP is assumed to be static.
- In the case of W-5GAN access: The User Location Information for W-5GAN is defined in TS 23.316 [84].
When the SMF receives a request to provide Access Network Information reporting while there is no action to carry out towards the 5G-AN or the UE (e.g. no QoS Flow to create / Update / modify), the SMF may request User Location Information from the AMF.
The interaction between AMF and SMF(s) for the case of a I-SMF insertion, relocation or removal for a PDU session is described in clause 5.34. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.3 Roaming | In the case of roaming the 5GC supports following possible deployments scenarios for a PDU Session:
- "Local Break Out" (LBO) where the SMF and all UPF(s) involved by the PDU Session are under control of the VPLMN.
- "Home Routed" (HR) where the PDU Session is supported by a SMF function under control of the HPLMN, by a SMF function under control of the VPLMN, by at least one UPF under control of the HPLMN and by at least one UPF under control of the VPLMN. In this case the SMF in HPLMN selects the UPF(s) in the HPLMN and the SMF in VPLMN selects the UPF(s) in the VPLMN. This is further described in clause 6.3. Home Routed with Session Breakout in VPLMN (HR-SBO) is described in clause 6.7 of TS 23.548 [130].
NOTE 1: The use of an UPF in the VPLMN e.g. enables VPLMN charging, VPLMN LI and minimizes the impact on the HPLMN of the UE mobility within the VPLMN (e.g. for scenarios where SSC mode 1 applies).
Different simultaneous PDU Sessions of an UE may use different modes: Home Routed and LBO. The HPLMN can control via subscription data per DNN and per S-NSSAI whether a PDU Session is to be set-up in HR or in LBO mode.
In the case of PDU Sessions per Home Routed deployment:
- NAS SM terminates in the SMF in VPLMN.
- The SMF in VPLMN forwards to the SMF in the HPLMN SM related information.
- The SMF in the HPLMN receives the SUPI of the UE from the SMF in the VPLMN during the PDU Session Establishment procedure.
- The SMF in HPLMN is responsible to check the UE request with regard to the user subscription and to possibly reject the UE request in the case of mismatch. The SMF in HPLMN obtains subscription data directly from the UDM.
- The SMF in HPLMN may send QoS requirements associated with a PDU Session to the SMF in VPLMN. This may happen during the PDU Session Establishment procedure and after the PDU Session is established. The interface between SMF in HPLMN and SMF in VPLMN is also able to carry (N9) User Plane forwarding information exchanged between SMF in HPLMN and SMF in VPLMN. The SMF in the VPLMN may check QoS requests from the SMF in HPLMN with respect to roaming agreements.
In home routed roaming case, the AMF selects an SMF in the VPLMN and a SMF in the HPLMN as described in clause 6.3.2 and in clause 4.3.2.2.3.3 of TS 23.502 [3] and provides the identifier of the selected SMF in the HPLMN to the selected SMF in the VPLMN.
In roaming with LBO, the AMF selects a SMF in the VPLMN as described in clause 6.3.2 and in clause 4.3.2.2.3.2 of TS 23.502 [3]. In this case, when handling a PDU Session Establishment Request message, the SMF in the VPLMN may reject the N11 message (related with the PDU Session Establishment Request message) with a proper N11 cause. This triggers the AMF to select both a new SMF in the VPLMN and a SMF in the HPLMN in order to handle the PDU Session using home routed roaming. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.4 Single PDU Session with multiple PDU Session Anchors | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.4.1 General | In order to support selective traffic routing to the DN or to support SSC mode 3 as defined in clause 5.6.9.2.3, the SMF may control the data path of a PDU Session so that the PDU Session may simultaneously correspond to multiple N6 interfaces. The UPF that terminates each of these interfaces is said to support PDU Session Anchor functionality. Each PDU Session Anchor supporting a PDU Session provides a different access to the same DN. Further, the PDU Session Anchor assigned at PDU Session Establishment is associated with the SSC mode of the PDU Session and the additional PDU Session Anchor(s) assigned within the same PDU Session e.g. for selective traffic routing to the DN are independent of the SSC mode of the PDU Session. When a PCC rule including the Application Function influence on traffic routing Enforcement Control information defined in clause 6.3.1 of TS 23.503 [45] is provided to the SMF, the SMF can decide whether to apply traffic routing (by using UL Classifier functionality or IPv6 multi-homing) based on DNAI(s) included in the PCC rule.
NOTE 1: Application Function influence on traffic routing Enforcement Control information can be either determined by the PCF when requested by AF via NEF as described in clause 5.6.7.1 or statically pre-configured in the PCF.
NOTE 2: Selective traffic routing to the DN supports, for example, deployments where some selected traffic is forwarded on an N6 interface to the DN that is "close" to the AN serving the UE.
This may correspond to
- The Usage of UL Classifier functionality for a PDU Session defined in clause 5.6.4.2.
- The Usage of an IPv6 multi-homing for a PDU Session defined in clause 5.6.4.3.
SMF may also take decision to apply traffic routing (by using UL Classifier functionality or IPv6 multi-homing) in EAS Discovery with EASDF procedure described in TS 23.548 [130]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.4.2 Usage of an UL Classifier for a PDU Session | In the case of PDU Sessions of type IPv4 or IPv6 or IPv4v6 or Ethernet, the SMF may decide to insert in the data path of a PDU Session an "UL CL" (Uplink classifier). The UL CL is a functionality supported by an UPF that aims at diverting (locally) some traffic matching traffic filters provided by the SMF. The insertion and removal of an UL CL is decided by the SMF and controlled by the SMF using generic N4 and UPF capabilities. The SMF may decide to insert in the data path of a PDU Session a UPF supporting the UL CL functionality during or after the PDU Session Establishment, or to remove from the data path of a PDU Session a UPF supporting the UL CL functionality after the PDU Session Establishment. The SMF may include more than one UPF supporting the UL CL functionality in the data path of a PDU Session.
The UE is unaware of the traffic diversion by the UL CL and does not involve in both the insertion and the removal of UL CL. In the case of a PDU Session of IPv4 or IPv6 or IPv4v6 type, the UE associates the PDU Session with either a single IPv4 address or a single IPv6 Prefix or both of them allocated by the network.
When an UL CL functionality has been inserted in the data path of a PDU Session, there are multiple PDU Session Anchors for this PDU Session. These PDU Session Anchors provide different access to the same DN. In the case of a PDU Session of IPv4 or IPv6 or IPv4v6 type, only one IPv4 address and/or IPv6 prefix is provided to the UE. The SMF may be configured with local policies for some (DNN, S-NSSAI) combinations to release the PDU Session when there is a PSA associated with the IPv4 address allocated to the UE and this PSA has been removed.
NOTE 0: The use of only one IPv4 address and/or IPv6 prefix with multiple PDU Session Anchors assumes that when needed, appropriate mechanisms are in place to correctly forward packets on the N6 reference point. The mechanisms for packet forwarding on the N6 reference point between the PDU Session Anchor providing local access and the DN are outside the scope of this specification.
The UL CL provides forwarding of UL traffic towards different PDU Session Anchors and merge of DL traffic to the UE i.e. merging the traffic from the different PDU Session Anchors on the link towards the UE. This is based on traffic detection and traffic forwarding rules provided by the SMF.
The UL CL applies filtering rules (e.g. to examine the destination IP address/Prefix of UL IP packets sent by the UE) and determines how the packet should be routed. The UPF supporting an UL CL may also be controlled by the SMF to support traffic measurement for charging, traffic replication for LI and bit rate enforcement (Session-AMBR per PDU Session).
NOTE 1: When N9 forwarding tunnel exists between source ULCL and target ULCL, the Session-AMBR per PDU Session can be enforced by the source UL CL UPF.
NOTE 2: The UPF supporting an UL CL may also support a PDU Session Anchor for connectivity to the local access to the data network (including e.g. support of tunnelling or NAT on N6). This is controlled by the SMF.
Additional UL CLs (and thus additional PDU Session Anchors) can be inserted in the data path of a PDU Session to create new data paths for the same PDU Session. The way to organize the data path of all UL CLs in a PDU Session is up to operator configuration and SMF logic and there is only one UPF supporting UL CL connecting to the (R)AN via N3 interface, except when session continuity upon UL CL relocation is used.
The insertion of an ULCL in the data path of a PDU Session is depicted in Figure 5.6.4.2-1.
Figure 5.6.4.2-1: User plane Architecture for the Uplink Classifier
NOTE 3: It is possible for a given UPF to support both the UL CL and the PDU Session Anchor functionalities.
Due to UE mobility the network may need to relocate the UPF acting as UL CL and establish a new PSA for local access to the DN. To support session continuity during UL CL relocation the network may establish a temporary N9 forwarding tunnel between the source UL CL and target UL CL. The AF may influence the creation of the N9 forwarding tunnel as described in clause 5.6.7.1.
The N9 forwarding tunnel is maintained until:
- all active traffic flowing on it ceases to exist for:
- a configurable period of time; or
- a period of time indicated by the AF;
- until the AF informs the SMF that it can release the source PSA providing local access to the DN.
During the existence of the N9 forwarding tunnel the UPF acting as target UL CL is configured with packet filters that:
- force uplink traffic from existing data sessions between UE and the application in the source local part of the DN (as defined in TS 23.548 [130]) into the N9 forwarding tunnel towards the source UL CL.
- force any traffic related to the application in the target local part of the DN to go to the new local part of the DN via the target PSA.
SMF may send a Late Notification to AF to inform it about the DNAI change as described in clause 4.3.6.3 of TS 23.502 [3]. This notification can be used by the AF e.g. to trigger mechanisms in the source local part of the DN to redirect the ongoing traffic sessions towards an application in the target local part of the DN. SMF can also send late notification to the target AF instance if associated with this target local part of the DN.
The procedure for session continuity upon UL CL relocation is described in clause 4.3.5.7 of TS 23.502 [3].
When an I-SMF is inserted for a PDU Session, the details of UL CL insertion which is controlled by an I-SMF is described in clause 5.34.4. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.4.3 Usage of IPv6 multi-homing for a PDU Session | A PDU Session may be associated with multiple IPv6 prefixes. This is referred to as multi-homed PDU Session. The multi-homed PDU Session provides access to the Data Network via more than one PDU Session Anchor. The different user plane paths leading to the different PDU Session Anchors branch out at a "common" UPF referred to as a UPF supporting "Branching Point" functionality. The Branching Point provides forwarding of UL traffic towards the different PDU Session Anchors and merge of DL traffic to the UE i.e. merging the traffic from the different PDU Session Anchors on the link towards the UE.
The UPF supporting a Branching Point functionality may also be controlled by the SMF to support traffic measurement for charging, traffic replication for LI and bit rate enforcement (Session-AMBR per PDU Session). The insertion and removal of a UPF supporting Branching Point is decided by the SMF and controlled by the SMF using generic N4 and UPF capabilities. The SMF may decide to insert in the data path of a PDU Session a UPF supporting the Branching Point functionality during or after the PDU Session Establishment, or to remove from the data path of a PDU Session a UPF supporting the Branching Point functionality after the PDU Session Establishment.
Multi homing of a PDU Session applies only for PDU Sessions of IPv6 type. When the UE requests a PDU Session of type "IPv4v6" or "IPv6" the UE also provides an indication to the network whether it supports a Multi-homed IPv6 PDU Session.
The use of multiple IPv6 prefixes in a PDU Session is characterised by the following:
- The UPF supporting a Branching Point functionality is configured by the SMF to spread UL traffic between the PDU Session Anchors based on the Source Prefix of the PDU (which may be selected by the UE based on routing information and preferences received from the network).
- IETF RFC 4191 [8] is used to configure routing information and preferences into the UE to influence the selection of the source Prefix.
NOTE 1: This corresponds to Scenario 1 defined in IETF RFC 7157 [7] "IPv6 Multi-homing without Network Address Translation". This allows to make the Branching Point unaware of the routing tables in the Data Network and to keep the first hop router function in the PDU Session Anchors.
- The multi-homed PDU Session may be used to support make-before-break service continuity to support SSC mode 3. This is illustrated in Figure 5.6.4.3-1.
- The multi-homed PDU Session may also be used to support cases where UE needs to access both a local service (e.g. local server) and a central service (e.g. the internet), illustrated in Figure 5.6.4.3-2.
- The UE shall use the method specified in clause 4.3.5.3 of TS 23.502 [3] to determine if a multi-homed PDU Session is used to support the service continuity case shown in Figure 5.6.4.3-1, or if it is used to support the local access to DN case shown in Figure 5.6.4.3-2.
Figure 5.6.4.3-1: Multi-homed PDU Session: service continuity case
NOTE 2: It is possible for a given UPF to support both the Branching Point and the PDU Session Anchor functionalities.
Figure 5.6.4.3-2: Multi-homed PDU Session: local access to same DN
NOTE 3: It is possible for a given UPF to support both the Branching Point and the PDU Session Anchor functionalities. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.5 Support for Local Area Data Network | The access to a DN via a PDU Session for a LADN is only available in a specific LADN service area. A LADN service area is a set of Tracking Areas. LADN is a service provided by the serving PLMN or the serving SNPN. It includes:
- LADN service applies only to 3GPP accesses and does not apply in Home Routed case.
- The usage of LADN DNN requires an explicit subscription to this DNN or subscription to a wildcard DNN.
- Whether a DNN corresponds to a LADN service is an attribute of a DNN and is per PLMN or per SNPN.
The UE is configured to know whether a DNN is a LADN DNN on a per-PLMN or per SNPN basis and an association between application and LADN DNN. The configured association is considered to be a UE local configuration defined in TS 23.503 [45]. Alternatively, the UE gets the information whether a DNN is a LADN DNN from LADN Information during (re‑)registration procedure as described in this clause.
NOTE 1: No other procedure for configuring the UE to know whether a DNN is a LADN DNN is defined in this release of the specifications.
NOTE 2: The procedure for configuring the UE to know an association between application and LADN DNN is not defined in this release of the specifications.
LADN service area and LADN DNN are configured in the AMF on a per DN basis, i.e. for different UEs accessing the same LADN, the configured LADN service area is the same regardless of other factors (e.g. UE's Registration Area or UE subscription).
NOTE 3: If a LADN is not available in any TA of an AMF's service area, the AMF is not required to be configured with any LADN related information for that DNN.
LADN Information (i.e. LADN Service Area Information and LADN DNN) is provided by AMF to the UE during the Registration procedure or UE Configuration Update procedure. For each LADN DNN configured in the AMF, the corresponding LADN Service Area Information includes a set of Tracking Areas that belong to the Registration Area that the AMF assigns to the UE (i.e. the intersection of the LADN service area and the assigned Registration Area). The AMF shall not create Registration Area based on the availability of LADNs.
NOTE 4: It is thus possible that the LADN Service Area Information sent by the AMF to the UE contains only a sub-set of the full LADN service area as the LADN service area can contain TA(s) outside of the registration area of the UE or outside of the area served by the AMF.
When the UE performs a successful (re-)registration procedure, the AMF may provide to the UE, based on local configuration (e.g. via OAM) about LADN, on UE location and on UE subscription information received from the UDM about subscribed DNN(s), the LADN Information for the list of LADN available to the UE in that Registration Area in the Registration Accept message.
The UE may provide either the LADN DNN(s) to retrieve the LADN Information for the indicated LADN DNN(s) or an indication of Requesting LADN Information to retrieve the LADN Information for all LADN(s) available in the current Registration Area.
The list of LADN is determined as follows:
- If neither LADN DNN nor an indication of requesting LADN Information is provided in the Registration Request message, the list of LADN is the LADN DNN(s) in subscribed DNN list except for wildcard DNN.
- If the UE provides LADN DNN(s) in the Registration Request message, the list of LADN is LADN DNN(s) the UE requested if the UE subscribed DNN(s) includes the requested LADN DNN or if a wildcard DNN is included in the UE's subscription data.
NOTE 5: It is assumed that an application can use only one LADN DNN at a time.
- If the UE provides an indication of requesting LADN Information in the Registration Request message, the list of LADN is all the LADN DNN(s) configured in the AMF if the wildcard DNN is subscribed, or the LADN DNN(s) which is in subscribed DNN list and no wildcard DNN is subscribed.
The UE considers the retrieved LADN Information valid only for the registered PLMN and the E-PLMN(s) if the LADN Service Area Information includes Tracking Areas that belong to E-PLMN(s). Additionally, an LADN DNN discovered by the UE via the retrieved LADN Information is considered an LADN DNN also in the E-PLMNs of the Registered PLMN, i.e. the UE can request LADN Information for the discovered LADN DNN in the E-PLMNs.
During the subsequent Registration procedure, if the network does not provide LADN Information for a DNN, the UE deletes any LADN Information for that DNN.
When the LADN Information for the UE in the 5GC is changed, the AMF shall update LADN Information to the UE through UE Configuration Update/Registration procedure as described in clauses 4.2.4 / 4.2.2.2.2 of TS 23.502 [3].
When receiving PDU Session Establishment with LADN DNN or Service Request for the established PDU Session corresponding to LADN, the AMF determines UE presence in LADN service area and forwards it to the SMF if the requested DNN is configured at the AMF as a LADN DNN.
Based on the LADN Service Area Information in the UE, the UE determines whether it is in or out of a LADN service area. If the UE does not have the LADN Service Area Information for a LADN DNN, the UE shall consider it is out of the LADN service area.
The UE takes actions as follows:
a) When the UE is out of a LADN service area, the UE:
- shall not request to activate UP connection of a PDU Session for this LADN DNN;
- shall not attempt to send user data as payload of a NAS message (see clause 5.31.4.1) using a PDU Session for this LADN DNN;
- shall not establish/modify a PDU Session for this LADN DNN (except for PS Data Off status change reporting for an established PDU Session);
- need not release any existing PDU Session for this LADN DNN unless UE receives explicit SM PDU Session Release Request message from the network.
b) When the UE is in a LADN service area, the UE:
- may request a PDU Session Establishment/Modification for this LADN DNN;
- may request to activate UP connection of the existing PDU Session for this LADN DNN;
- may attempt to send user data as payload of a NAS message (see clause 5.31.4.1) using a PDU Session for this LADN DNN.
NOTE 6: The evaluation of Service Area Restrictions will be performed before the evaluation of LADN service area, if the UE has overlapping areas between Service Area Restrictions and LADN service area.
The SMF supporting a DNN is configured with information about whether this DNN is a LADN DNN or not.
When receiving SM request corresponding an LADN from the AMF, the SMF determines whether the UE is inside LADN service area based on the indication (i.e. UE Presence in LADN service area) received from the AMF. If the SMF does not receive the indication, the SMF considers that the UE is outside of the LADN service area. The SMF shall reject the request if the UE is outside of the LADN service area.
When the SMF receives a request for PDU Session Establishment with the LADN DNN, it shall subscribe to "UE mobility event notification" for reporting UE presence in Area of Interest by providing LADN DNN to the AMF as described in clauses 5.6.11 and 5.3.4.4.
Based on the notification about the UE presence in LADN service area notified by AMF (i.e. IN, OUT, or UNKNOWN), the SMF takes actions as follows based on operator's policy:
a) When SMF is informed that the UE presence in a LADN service area is OUT, the SMF shall:
- release the PDU Session immediately; or
- deactivate the user plane connection for the PDU Session and it shall not attempt to send user data as payload of a NAS message (see clause 5.31.4.1) while maintaining the PDU Session and ensure the Data Notification is disabled and the SMF may release the PDU Session if the SMF is not informed that the UE moves into the LADN service area after a period.
b) When SMF is informed that the UE presence a LADN service area is IN, the SMF shall:
- ensure that Data Notification is enabled.
- trigger the Network triggered Service Request procedure for a LADN PDU Session to active the UP connection or send user data as payload of a NAS message (see clause 5.31.4.1) when the SMF receives downlink data or Data Notification from UPF.
c) When the SMF is informed that the UE presence in a LADN service area is UNKNOWN, the SMF may:
- ensure that Data Notification is enabled.
- trigger the Network triggered Service Request procedure for a LADN PDU Session to active the UP connection or send user data as payload of a NAS message (see clause 5.31.4.1) when the SMF receives downlink data or Data Notification from UPF.
SMF may make use of UE mobility analytics provided by NWDAF, as described in clause 6.7.2 of TS 23.288 [86], to determine UE presence pattern in LADN service area and take a decision how to handle LADN PDN Session (e.g. release the PDU Session immediately, or deactivate the user plane connection for the PDU Session with maintaining the PDU Session).
5.6.5a Supporting LADN per DNN and S-NSSAI
The 5GS may support LADN per DNN and S-NSSAI and the functions specified in clause 5.6.5 are used (with the extension to apply per DNN and S-NSSAI whenever applicable) with the following enhancements:
- The UE indicates the support of LADN per DNN and S-NSSAI in the UE MM Core Network Capability of the Registration Request message.
- The LADN service area can be provisioned for a group (e.g. 5G VN group) or an individual subscriber using UDM/NEF parameter provisioning service triggered by AF request as described in clause 4.15.6 of TS 23.502 [3].
- LADN service area per DNN and S-NSSAI may be configured in the AMF. The LADN service area may also be provided to AMF as part of the subscription data from UDM.
- The LADN Service Area Information provided to the UE is determined by AMF, based on the Registration Area that the AMF assigns to the UE and either the local configured LADN service area or the LADN service area received from UDM. In case there is both a configured LADN service area and a LADN service area received from UDM, the AMF decides based on operator configuration which one takes precedence.
- If the UE indicates support of LADN per DNN and S-NSSAI, the AMF may provide the LADN Service Area Information per LADN DNN and S-NSSAI to the UE.
- When the UE has an ongoing PDU session and there is no LADN service area for the DNN and S-NSSAI of the PDU session, the AMF will release the ongoing PDU session if the AMF determines to configure the LADN service area configured per LADN DNN and S-NSSAI for the associated DNN and S-NSSAI (e.g. due to notification from UDM or local configuration update).
- When the UE has an ongoing PDU session subject to LADN per LADN DNN and S-NSSAI, the AMF will release the ongoing PDU session if the AMF determines the LADN service area for the DNN and S-NSSAI is removed (e.g. due to notification from UDM or local configuration update).
NOTE 1: After the UE receives the LADN Information per LADN DNN and S-NSSAI, it is up to the UE implementation whether to re-establish the PDU session.
- If the UE does not indicate support of LADN per DNN and S-NSSAI and the AMF has a LADN service area for the DNN as well as a LADN service area for the DNN and S-NSSAI, the AMF may determine the LADN Service Area Information per LADN DNN sent to UE using the LADN service area for the DNN as described in clause 5.6.5.
- If the UE does not indicate support of LADN per DNN and S-NSSAI and the AMF has no LADN service area for the DNN but there is a LADN service area for the DNN and S-NSSAI, then the AMF may determine the LADN Service Area Information per LADN DNN sent to UE using this LADN service area for the DNN and S-NSSAI as the LADN service area for that DNN as described in clause 5.6.5 if the UE is not subscribed to the same DNN for multiple S-NSSAI(s) (i.e. the UE is subscribed to the DNN for a single S-NSSAI only).
NOTE 2: In order to serve the UEs that do not support LADN per DNN and S-NSSAIs, the operator can avoid using the same DNN for multiple S-NSSAIs if LADN service area is configured per DNN and S-NSSAI.
- If the UE does not indicate support of LADN per DNN and S-NSSAI and the AMF neither has a LADN service area for the DNN nor has a LADN service area for the DNN and S-NSSAI, the AMF shall not provide any LADN Information to the UE.
- The AMF enforces the LADN Service Area per LADN DNN and S-NSSAI in the same way as is described in clause 5.6.5 with the difference that the LADN service area is defined per DNN and S-NSSAI.
- The UE enforces the LADN Service Area per LADN DNN and S-NSSAI, if received from the AMF, in the same way as is described in clause 5.6.5 with the difference that the LADN service area is defined per DNN and S-NSSAI.
- If the AMF enforces the LADN Service Area per LADN DNN and S-NSSAI for the UE, the AMF indicates to the SMF during PDU Session Establishment that the PDU Session is subject to LADN per LADN DNN and S-NSSAI.
- If indicated by AMF at PDU Session Establishment that PDU Session is subject to LADN per LADN DNN and S-NSSAI, the SMF configures the DNN of PDU Session as LADN DNN. The SMF shall subscribe to "UE mobility event notification" for reporting UE presence in Area of Interest by providing LADN DNN and S-NSSAI to the AMF as described in clauses 5.6.11 and 5.3.4.4. The SMF handles the PDU session in the same way as described in clause 5.6.5.
NOTE 3: If the UE has overlapping areas between LADN service area configured per DNN & S-NSSAI, Partial network slice support area, or any combination of them, then the evaluation of Partial network slice support area take precedence over the evaluation of LADN service area configured per DNN and S-NSSAI. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.6 Secondary authentication/authorization by a DN-AAA server during the establishment of a PDU Session | At PDU Session Establishment to a DN:
- The DN-specific identity (TS 33.501 [29]) of a UE may be authenticated/authorized by the DN.
NOTE 1: the DN-AAA server may belong to the 5GC or to the DN.
- If the UE provides authentication/authorization information corresponding to a DN-specific identity during the Establishment of the PDU Session and the SMF determines that Secondary authentication/authorization of the PDU Session Establishment is required based on the SMF policy associated with the DN, the SMF passes the authentication/authorization information of the UE to the DN-AAA server via the UPF if the DN-AAA server is located in the DN. If the SMF determines that Secondary authentication/authorization of the PDU Session Establishment is required but the UE has not provided a DN-specific identity as part of the PDU Session Establishment request, the SMF requests the UE to indicate a DN-specific identity using EAP procedures as described in TS 33.501 [29]. If the Secondary authentication/authorization of the PDU Session Establishment fails, the SMF rejects the PDU Session Establishment.
NOTE 2: If the DN-AAA server is located in the 5GC and reachable directly, then the SMF may communicate with it directly without involving the UPF.
- The DN-AAA server may authenticate/authorize the PDU Session Establishment.
- When DN-AAA server authorizes the PDU Session Establishment, it may send DN Authorization Data for the established PDU Session to the SMF. The DN authorization data for the established PDU Session may include one or more of the following:
- A DN Authorization Profile Index which is a reference to authorization data for policy and charging control locally configured in the SMF or PCF.
- a list of allowed MAC addresses for the PDU Session; this shall apply only for PDU Session of Ethernet PDU type and is further described in clause 5.6.10.2.
- a list of allowed VLAN tags for the PDU Session; this shall apply only for PDU Session of Ethernet PDU type and is further described in clause 5.6.10.2.
- VLAN handling information (e.g. the C-TAG to be inserted or removed, S-TAG to be inserted or removed); this shall apply only for PDU Session of Ethernet PDU type and is further described in clause 5.6.10.2.
- DN authorized Session AMBR for the PDU Session. The DN Authorized Session AMBR for the PDU Session takes precedence over the subscribed Session-AMBR received from the UDM.
- Framed Route information (see clause 5.6.14) for the PDU Session.
- L2TP information, such as LNS IP address and/or LNS host name, as described in TS 29.561 [132].
SMF policies may require DN authorization without Secondary authentication/authorization. In that case, when contacting the DN-AAA server for authorization, the SMF provides the GPSI of the UE if available.
Such Secondary authentication/authorization takes place for the purpose of PDU Session authorization in addition to:
- The 5GC access authentication handled by AMF and described in clause 5.2.
- The PDU Session authorization enforced by SMF with regards to subscription data retrieved from UDM.
Based on local policies the SMF may initiate Secondary authentication/authorization at PDU Session Establishment. The SMF provides the GPSI, if available, in the signalling exchanged with the DN-AAA during Secondary authentication/authorization.
After the successful Secondary authentication/authorization, a session is kept between the SMF and the DN-AAA.
The UE provides the authentication/authorization information required to support Secondary authentication/authorization by the DN over NAS SM.
If a UE is configured with DNNs, which are subject to secondary authentication/authorization, the UE stores an association between the DNN and corresponding credentials for the secondary authentication/authorization.
NOTE 3: How the UE is aware that a DNN is subject to secondary authentication/authorization (e.g. based on local configuration) is out of scope of this specification.
The UE may support remote provisioning of credentials for secondary authentication/authorization, as specified in clause 5.39.
A UE that supports to be provisioned with the credentials used for secondary authentication/authorization over UP remote provisioning shall use connectivity over an S-NSSAI/DNN which can access the provisioning server to establish a PDU session for remote provisioning as defined in clause 5.39.
NOTE 4: The credentials for secondary authentication/authorization are not specified.
SMF policies or subscription information (such as defined in Table 5.2.3.3.1 of TS 23.502 [3]) may trigger the need for SMF to request the Secondary authentication/authorization and/or UE IP address / Prefix from the DN-AAA server.
When SMF adds a PDU Session Anchor (such as defined in clause 5.6.4) to a PDU Session Secondary authentication/authorization is not carried out, but SMF policies may require SMF to notify the DN when a new prefix or address has been added to or removed from a PDU Session or N6 traffic routing information has been changed for a PDU Session.
When SMF gets notified from UPF with the addition or removal of MAC addresses to/from a PDU Session, the SMF policies may require SMF to notify the DN-AAA server.
Indication of PDU Session Establishment rejection is transferred by SMF to the UE via NAS SM.
If the DN-AAA sends DN Authorization Data for the authorized PDU Session to the SMF and dynamic PCC is deployed, the SMF sends the PCF the DN authorized Session AMBR and/or DN Authorization Profile Index in the DN Authorization Data for the established PDU Session.
If the DN-AAA sends DN Authorization Profile Index in DN Authorization Data to the SMF and dynamic PCC is not deployed, the SMF uses the DN Authorization Profile Index to refer the locally configured information.
NOTE 5: DN Authorization Profile Index is assumed to be pre-negotiated between the operator and the administrator of DN-AAA server.
If the DN-AAA does not send DN Authorization Data for the established PDU Session, the SMF may use locally configured information.
At any time, a DN-AAA server may revoke the authorization for a PDU Session or update DN Authorization Data for a PDU Session. According to the request from DN-AAA server, the SMF may release or update the PDU Session. See clause 5.6.14 when the update involves Framed Route information.
At any time, a DN-AAA server or SMF may trigger Secondary Re-authentication procedure for a PDU Session established with Secondary Authentication as specified in clause 11.1.3 of TS 33.501 [29].
During Secondary Re-authentication/Re-authorization, if the SMF receives from DN-AAA the DN authorized Session AMBR and/or DN Authorization Profile Index, the SMF shall report the received value(s) to the PCF.
The procedure for secondary authentication/authorization by a DN-AAA server during the establishment of a PDU Session is described in clause 4.3.2.3 of TS 23.502 [3].
The support for L2TP on N6 is further specified in clause 5.8.2.16 and the procedure for establishment of L2TP tunnelling on N6 for a PDU Session is described in clause 4.3.2.4 of TS 23.502 [3].
NOTE 6: The L2TP Tunnel information sent to the SMF can, for example, be provisioned in the DN-AAA server per DNN/S-NSSAI or per SUPI or GPSI. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.7 Application Function influence on traffic routing | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.7.1 General | The content of this clause applies to non-roaming and to LBO deployments i.e. to cases where the involved entities (AF, PCF, SMF, UPF) belong to the Serving PLMN or AF belongs to a third party with which the Serving PLMN has an SLA agreement.
AF influence on traffic routing may apply in the case of Home Routed deployments with Session Breakout (HR SBO) as defined in TS 23.548 [130]. In that case when an AF belonging to the V-PLMN (or with an offloading SLA with the V PLMN) desires to provide Traffic Influence policies it may invoke at the V-NEF the API defined in this clause and provide the information listed in Table 5.6.7-1 below but the corresponding Traffic Influence information is provided directly from V-NEF to V-SMF bypassing the PCF. This is further defined in TS 23.548 [130] clause 6.7.2. and the rest of the clause 5.6.7.1 does not address how information related with AF influence on traffic routing may be provided to the SMF in the case of HR SBO.
PCF shall not apply AF requests to influence traffic routing to PDU Sessions established in Home Routed mode.
The AF may determine the common EAS/DNAI for the UE set in order to indicate a common EAS or common local part of DN and provide the common EAS/DNAI to the 5GS.
An AF may send requests to influence SMF routing decisions for traffic of PDU Session. The AF requests may influence UPF (re)selection and (I-)SMF (re)selection and allow routing user traffic to a local access to a Data Network (identified by a DNAI).
The AF may issue requests on behalf of applications not owned by the PLMN serving the UE.
If the operator does not allow an AF to access the network directly, the AF shall use the NEF to interact with the 5GC, as described in clause 6.2.10.
The AF may be in charge of the (re)selection or relocation of the applications within the local part of the DN (as defined in TS 23.548 [130]). Such functionality is not defined. For this purpose, the AF may request to get notified about events related with PDU Sessions.
In the case of AF instance change, the AF may send request of AF relocation information.
The AF requests are sent to the PCF via N5 (in the case of requests targeting specific on-going PDU Sessions of individual UE(s), for an AF allowed to interact directly with the 5GC NFs) or via the NEF. The AF requests that target existing or future PDU Sessions of multiple UE(s) or of any UE are sent via the NEF and may target multiple PCF(s), as described in clause 6.3.7.2. The PCF(s) transform(s) the AF requests into policies that apply to PDU Sessions. When the AF has subscribed to UP path management event notifications from SMF(s) (including notifications on how to reach a GPSI over N6), such notifications are sent either directly to the AF or via an NEF (without involving the PCF). For AF interacting with PCF directly or via NEF, the AF requests may contain the information as described in the Table 5.6.7-1:
Table 5.6.7-1: Information element contained in AF request
Information Name
Applicable for PCF or NEF (NOTE 1)
Applicable for NEF only
Category
Traffic Description
Defines the target traffic to be influenced, represented by the combination of DNN and optionally S-NSSAI and optionally PLMN ID of the PLMN that the DNN/S-NSSAI belongs to and application identifier or traffic filtering information. (NOTE 7) (NOTE 8)
The target traffic can be represented by AF-Service-Identifier, instead of combination of DNN and optionally S-NSSAI.
Mandatory
Potential Locations of Applications
Indicates potential locations of applications, represented by a list of DNAI(s) and optionally PLMN ID of the PLMN that the list of DNAI(s) belongs to. (NOTE 8)
The potential locations of applications can be represented by AF-Service-Identifier.
Conditional
(NOTE 2)
Target UE Identifier(s)
Indicates the UE(s) that the request is targeting, i.e. one or a list of individual UE(s), a group of UE represented by Internal Group Identifier(s) (NOTE 3), or any UE accessing the combination of DNN, S-NSSAI and DNAI(s).
GPSI can be applied to identify the individual UE, or External Group Identifier(s) can be applied to identify a group of UE (NOTE 3). External Subscriber Category(s) (NOTE 5).
Mandatory
Spatial Validity Condition
Indicates that the request applies only to the traffic of UE(s) located in the specified location, represented by areas of validity.
The specified location can be represented by geographical area.
Optional
AF transaction identifier
The AF transaction identifier refers to the AF request.
N/A
Mandatory
N6 Traffic Routing requirements
Routing profile ID and/or N6 traffic routing information corresponding to each DNAI and an optional indication of traffic correlation (NOTE 4).
N/A
Optional
(NOTE 2)
Application Relocation Possibility
Indicates whether an application can be relocated once a location of the application is selected by the 5GC.
N/A
Optional
UE IP address preservation indication
Indicates UE IP address should be preserved.
N/A
Optional
Temporal Validity Condition
Time interval(s) or duration(s).
N/A
Optional
Information on AF subscription to corresponding SMF events
Indicates whether the AF subscribes to change of UP path of the PDU Session and the parameters of this subscription.
N/A
Optional
Information for EAS IP Replacement in 5GC
Indicates the Source EAS identifier and Target EAS identifier, (i.e. IP addresses and port numbers of the source and target EAS).
N/A
Optional
User Plane Latency Requirement
Indicates the user plane latency requirements
N/A
Optional
Information on AF change
N/A
Indicates the AF instance relocation and relocation information.
Optional
Indication for EAS Relocation
Indicates the EAS relocation of the application(s)
N/A
Optional
Indication for Simultaneous Connectivity over the source and target PSA at Edge Relocation
Indicates that simultaneous connectivity over the source and target PSA should be maintained at edge relocation and provides guidance to determine when the connectivity over the source PSA can be removed.
N/A
Optional
EAS Correlation indication
Indicates selecting a common EAS for the application identified by the Traffic Description for the set of UEs.
Optional
Common EAS IP address
the common EAS for the application identified by the Traffic Description for a set of UEs the AF request aims at.
Optional
Traffic Correlation ID
Identification of a set of UEs targeted at by the AF request and accessing the application identified by the Traffic Description.
Optional
FQDN(s)
FQDN(s) used for influencing EASDF-based DNS query procedure as defined in TS 23.548 [130] (NOTE 6).
Optional
Indication of considering N6 delay
Indicates that N6 delay should be considered (NOTE 9).
Optional
NOTE 1: When the AF request targets existing or future PDU Sessions of multiple UE(s) or of any UE and is sent via the NEF, as described in clause 6.3.7.2, the information is stored in the UDR by the NEF and notified to the PCF by the UDR.
NOTE 2: The potential locations of applications and N6 traffic routing requirements may be absent only if the request is for subscription to notifications about UP path management events or the request is for indication of selecting Common EAS for a set of UEs.
NOTE 3: Internal Group ID can only be used by an AF controlled by the operator and only towards PCF. If a list of Internal/External Group IDs is provided by the AF, the AF request applies to the UEs that belong to every one of these groups, i.e. a single UE needs to be a member of every group in the list of Internal/External Group IDs.
NOTE 4: The indication of traffic correlation can be used for 5G VN groups as described in clause 5.29.
NOTE 5: External Subscriber category(s) can be combined with External Group ID(s) or any UE. If a list of External Subscriber categories is provided by the AF, the AF request applies to the UEs that belong to every one of these Subscriber categories, i.e. the UE has all these Subscriber categories in its subscription data.
NOTE 6: FQDN(s) is used for influencing EASDF-based DNS query procedure as defined in clause 6.2.3.2.2 of TS 23.548 [130].
NOTE 7: If the FQDN is included in the AF request, the EASDF-based EAS discovery procedure will be followed as defined in TS 23.548 [130] using this FQDN for the purpose of setting traffic route and finding DNAI and Traffic Description will be ignored.
NOTE 8: PLMN ID is used for HR-SBO case as defined in clause 4.3.6.1 of TS 23.502 [3]. In this case, DNN and S-NSSAI are the ones of the HPLMN.
NOTE 9: When the Indication of considering N6 delay is provided, N6 delay measurement information as part of EDI, as described in Table 6.2.3.4.1 of TS 23.548 [130] needs to be used to perform measurements.
For each information element mentioned above in the AF request, the detailed description is as follows:
1) Information to identify the traffic. The traffic can be identified in the AF request by
- Either a DNN and possibly slicing information (S-NSSAI) or an AF-Service-Identifier
- When the AF provides an AF-Service-Identifier i.e. an identifier of the service on behalf of which the AF is issuing the request, the 5G Core maps this identifier into a target DNN and slicing information (S-NSSAI)
- When the NEF processes the AF request the AF-Service-Identifier may be used to authorize the AF request.
- An application identifier or traffic filtering information (e.g. IP 5 Tuple). The application identifier refers to an application handling UP traffic and is used by the UPF to detect the traffic of the application.
When the AF request is for influencing SMF routing decisions, the information is to identify the traffic to be routed.
When the AF request is for subscription to notifications about UP path management events, the information is to identify the traffic that the events relate to. The AF request may include a PLMN ID of the PLMN that the DNN and S-NSSAI belong to, as described in clause 4.3.6.1 of TS 23.502 [3].
2) Information about the N6 traffic routing requirements for traffic identified as defined in 1). This includes:
- Information about the N6 traffic routing requirements that is provided per DNAI: for each DNAI, the N6 traffic routing requirements may contain a routing profile ID and/or N6 traffic routing information.
- An optional indication of traffic correlation, when the information in 4) identifies a group of UEs. This implies the targeted PDU Sessions should be correlated by a common DNAI in the user plane for the traffic identified in 1). If this indication is provided by the AF, the 5GC should select a common DNAI for the target PDU Sessions from the list of DNAI(s) specified in 3).
NOTE 1: The N6 traffic routing requirements are related to the mechanism enabling traffic steering in the local access to the DN. The routing profile ID refers to a pre-agreed policy between the AF and the 5GC. This policy may refer to different steering policy ID(s) sent to SMF and e.g. based on time of the day etc.
NOTE 2: The mechanisms enabling traffic steering in the local access to the DN are not defined.
3) Potential locations of applications towards which the traffic routing should apply. The potential location of application is expressed as a list of DNAI(s). If the AF interacts with the PCF via the NEF, the NEF may map the AF-Service-Identifier information to a list of DNAI(s). The DNAI(s) may be used for UPF (re)selection and (I‑)SMF (re)selection. When only one DNAI is included and the Indication of traffic correlation is available, the DNAI is used as common DNAI for UEs identified by AF request. The AF request may include a PLMN ID of the PLMN that the list of DNAI(s) belongs to, as described in clause 4.3.6.1 of TS 23.502 [3].
4) Information on the set of target UE(s). This may correspond to:
- Individual UEs (i.e. one or a list of UEs) identified using GPSI, or an IP address/Prefix or a MAC address.
- Group(s) of UEs identified by External Group Identifier(s) as defined in TS 23.682 [36] when the AF interacts via the NEF, or Internal-Group Identifier (see clause 5.9.7) when the AF interacts directly with the PCF.
- Any UE accessing the combination of DNN, S-NSSAI and DNAI(s).
- External Group ID(s) or any UE can both be complemented with External Subscriber Category(s) for a more granular selection of UEs. NEF may map this to Internal Group ID(s) or a combination of Internal Group ID(s) and Subscriber Category(s), defined in TS 23.503 [45].
NOTE 3: Only NEF is aware of the External Subscriber Category. As a user can be associated with multiple Subscriber Category(s), some values of Subscriber Category(s) can correspond to an SLA between an application provider represented by an AF and the 5GC operator. In the NEF API, the combination of application identifier and External Subscriber Category can also be used to refer to this SLA.
When the PDU Session type is IPv4 or IPv6 or IPv4v6 and the AF provides an IP address and/or an IP Prefix, or when the PDU Session type is Ethernet and the AF provides a MAC address, this allows the PCF to identify the PDU Session for which this request applies and the AF request applies only to that specific PDU Session of the UE. In this case, additional information such as the UE identity may also be provided to help the PCF to identify the correct PDU Session.
Otherwise, the request targets multiple UE(s) and shall apply to any existing or future PDU Sessions that match the parameters in the AF request.
When the AF request targets an individual or a list of UE(s) and GPSI is provided within the AF request, the GPSI is mapped to SUPI according to the subscription information received from UDM.
When the AF request targets any UE or a group of UE, the AF request is likely to influence multiple PDU Sessions possibly served by multiple SMFs and PCFs.
When the AF request targets a group of UE it provides one or several group identifiers in its request. The group identifiers provided by the AF are mapped to Internal-Group identifiers. Members of the group have Group Identifier(s) in their subscription. The Internal-Group Identifier(s) is(are) stored in UDM, retrieved by SMF from UDM and passed by SMF to PCF at PDU Session set-up. The PCF can then map the AF request with user subscription and determine whether an AF request targeting a Group of users applies to a PDU Session. When External Subscriber Category(s) is provided, the NEF maps External Subscriber Category(s) into Subscriber Category(s), the PCF can map the AF request with user subscription and then creates PCC rules for UEs that have the Subscriber Category(s) in their subscription.
When the AF request is for influencing SMF routing decisions, the information is to identify UE(s) whose traffic is to be routed.
When the AF request is for subscription to notifications about UP path management events, the information is to identify UE(s) whose traffic the events relate to.
When the AF request is for traffic forwarding in a PDU Session serving for TSC, the MAC address used by the PDU Session is determined by the AF to identify UE whose traffic is to be routed according to the previously stored binding relationship of the 5GS Bridge and the port number of the traffic forwarding information received from TSN network.
5) Indication of application relocation possibility. This indicates whether an application can be relocated once a location of the application is selected by the 5GC. If application relocation is not possible, the 5GC shall ensure that for the traffic related with an application, no DNAI change takes place once selected for this application.
6) Temporal validity condition. This is provided in the form of time interval(s) or duration(s) during which the AF request is to be applied.
When the AF request is for influencing SMF routing decisions, the temporal validity condition indicates when the traffic routing is to apply.
When the AF request is for subscription to notifications about UP path management events, the temporal validity condition indicates when the notifications are to be generated.
7) Spatial validity condition on the UE(s) location. This is provided in the form of validity area(s). If the AF interacts with the PCF via the NEF, it may provide geographical area (e.g. a civic address or shapes) and the NEF maps the information to areas of validity based on pre-configuration. The PCF in turn determines area(s) of interest based on validity area(s).
When the AF request is for influencing SMF routing decisions, the spatial validity condition indicates that the request applies only to the traffic of UE(s) located in the specified location.
When the AF request is for subscription to notifications about UP path management events, the spatial validity condition indicates that the subscription applies only to the traffic of UE(s) located in the specified location.
8) Information on AF subscription to corresponding SMF events.
The AF may request to be subscribed to change of UP path associated with traffic identified in the bullet 1) above. This information element may also be used for subscribing to notification of the AF traffic influence result for "Simultaneous Connectivity over the source and target PSA at Edge Relocation" if it is provided together with the Indication for Simultaneous Connectivity over source and target PSA at Edge Relocation in the bullet 15).
The AF request contains:
- A type of subscription (subscription for Early and/or Late notifications).
The AF subscription can be for Early notifications and/or Late notifications. In the case of a subscription for Early notifications, the SMF sends the notifications before the (new) UP path is configured. In the case of a subscription for Late notifications, the SMF sends the notification after the (new) UP path has been configured.
- Notification target address for receiving event notification.
- Optionally, an indication of "AF acknowledgment to be expected".
The indication implies that the AF will provide a response to the notifications of UP path management events to the 5GC. The SMF may, according to this indication, determine to wait for a response from the AF before the SMF configures in the case of early notification, or activates in the case of late notification, the new UP path as described in clause 5.6.7.2.
- Optionally, an immediate reporting flag.
The immediate reporting flag is included when AF subscribe for candidate DNAI(s) of UE for common EAS/DNAI selection. With this flag, SMF should immediately response AF with the candidate DNAI(s) using Notification of user plane management event as described in clause 4.3.6.3 in TS 23.502 [3].
The AF subscription can also request to receive information associating the GPSI of the UE with the IP address(es) of the UE and/or with actual N6 traffic routing to be used to reach the UE on the PDU Session; in this case the corresponding information is sent by the SMF regardless of whether a DNAI applies to the PDU Session.
9) An AF transaction identifier referring to the AF request. This allows the AF to update or remove the AF request and to identify corresponding UP path management event notifications. The AF transaction identifier is generated by the AF.
When the AF interacts with the PCF via the NEF, the NEF maps the AF transaction identifier to an AF transaction internal identifier, which is generated by the NEF and used within the 5GC to identify the information associated to the AF request. The NEF maintains the mapping between the AF transaction identifier and the AF transaction internal identifier. The relation between the two identifiers is implementation specific.
When the AF interacts with the PCF directly, the AF transaction identifier provided by the AF is used as AF transaction internal identifier within the 5GC.
10) Indication of UE IP address preservation. This indicates UE IP address related to the traffic identified in bullet 1) should be preserved. If this indication is provided by the AF, the 5GC should preserve the UE IP address by preventing reselection of PSA UPF for the identified traffic once the PSA UPF is selected.
11) Information for EAS IP Replacement in 5GC. This indicates the Source EAS identifier and Target EAS identifier (i.e. IP addresses and port numbers of the source and target EAS) for a service subject to Edge Computing.
12) User Plane Latency Requirement. This includes AF requirements for User Plane latency. (see clause 6.3.6 of TS 23.548 [130]).
13) Information on AF change. The AF relocation information includes:
- AF Identifier: the identifier of the target AF instance.
NOTE 4: The AF relocation information is applicable for interaction with NEF only and it is not stored in UDR or transferred to PCF, even for the case AF directly interacts with PCF.
14) Indication for EAS relocation. This indicates the application(s) are to be relocated.
15) Indication for Simultaneous Connectivity over source and target PSA at Edge Relocation (see clause 6.3.4 of TS 23.548 [130]). Indicates that source and target PSA should coexist for some time at PSA relocation and may influence the establishment of a temporary N9 forwarding tunnel between the source UL CL and target UL CL. It may also provide guidance for the time interval after the described traffic ceases when the connectivity over the source PSA may be removed.
16) Traffic Correlation ID. Identification of a set of UEs subjecting to the AF request and accessing the application identified by the Traffic Description. UEs associated with the same Traffic Correlation ID and accessing the application identified by the Traffic Description should connect to a common EAS or EAS(es) corresponding to a common DNAI.
The following attributes may be provided with the Traffic Correlation ID:
- EAS Correlation indication. Indicates selecting a common EAS for a set of UEs identified by Traffic Correlation ID and accessing the application identified by the Traffic Description.
- Common EAS IP address. IP address of the common EAS to be accessed by the UEs in the set of UEs subject to AF request, for the application identified by the Traffic Description.
NOTE 5: In the case of common EAS selection, if Traffic Correlation ID is provided, either EAS Correlation indication or Common EAS will be included in AF request.
- FQDN(s). FQDN(s) corresponding to the application to be accessed by a set of UEs. When FQDN(s) is included, it is used for influencing EAS discovery procedure as defined in TS 23.548 [130].
- Indication of traffic correlation as described in 2).
NOTE 6: If both EAS Correlation indication and Indication of traffic correlation are included, the set of UEs access the common EAS via the common DNAI.
17) Indication of considering N6 delay. This is used to trigger the N6 delay measurement and indicate that N6 delay measurement, if available, should be considered by the SMF to (re)select the PSA UPF(s) or trigger EAS(es) (re)discovery.
An AF may send requests to influence SMF routing decisions, for event subscription or for both.
The AF may request to be subscribed to notifications about UP path management events, i.e. a UP path change occurs for the PDU Session. The corresponding notification about a UP path change sent by the SMF to the AF may indicate the DNAI and /or the N6 traffic routing information and/or common EAS that has changed as described in clause 4.3.6.3 of TS 23.502 [3]. It may include the AF transaction internal identifier, the type of notification (i.e. early notification or late notification), the Identity of the source and/or target DNAI, the IP address/prefix of the UE or the MAC address used by the UE, the GPSI and the N6 traffic routing information related to the 5GC UP. The AF may subscribe for notifications of candidate DNAI(s) for UE if AF selection of common EAS/DNAI for a set of UEs is used.
NOTE 7: The change from the UP path status where no DNAI applies to a status where a DNAI applies indicates the activation of this AF request; the change from the UP path status where a DNAI applies to a status where no DNAI applies indicates the de-activation of this AF request.
In the case of IP PDU Session Type, the IP address/prefix of the UE together with N6 traffic routing information indicates to the AF how to reach over the User Plane the UE identified by its GPSI. N6 traffic routing information indicates any tunnelling that may be used over N6. The nature of this information depends on the deployment.
NOTE 8: N6 traffic routing information can e.g. correspond to the identifier of a VPN or to explicit tunnelling information such as a tunnelling protocol identifier together with a Tunnel identifier.
NOTE 9: In the case of Unstructured PDU Session type the nature of the N6 traffic routing information related to the 5GC UP is described in clause 5.6.10.3.
In the case of Ethernet PDU Session Type, the MAC address of the UE together with N6 traffic routing information indicates to the AF how to reach over the User Plane the UE identified by its GPSI. The UE MAC address (es) is reported by the UPF as described in clause 5.8.2.12. The N6 traffic routing information can be, e.g. a VLAN ID or the identifier of a VPN or a tunnel identifier at the UPF.
When notifications about UP path management events are sent to the AF via the NEF, if required, the NEF maps the UE identify information, e.g. SUPI, to the GPSI and the AF transaction internal identifier to the AF transaction identifier before sending the notifications to the AF.
The PCF, based on information received from the AF, operator's policy, optionally service experience analytics per UP path received from NWDAF, etc. authorizes the request received from the AF and determines for each DNAI, a traffic steering policy ID (derived from the routing profile ID provided by the AF) and/or the N6 traffic routing information (as provided by the AF) to be sent to the SMF as part of the PCC rules. The traffic steering policy IDs are configured in the SMF or in the UPF. The traffic steering policy IDs are related to the mechanism enabling traffic steering to the DN.
The DNAIs are related to the information considered by the SMF for UPF selection and (I‑)SMF (re)selection, e.g. for diverting (locally) some traffic matching traffic filters provided by the PCF.
The PCF acknowledges a request targeting an individual PDU Session to the AF or to the NEF.
For PDU Session that corresponds to the AF request, the PCF provides the SMF with a PCC rule that is generated based on the AF request, Local routing indication from the PDU Session policy control subscription information and taking into account UE location presence in area of interest (i.e. Presence Reporting Area). The PCC rule contains the information to identify the traffic, information about the DNAI(s) towards which the traffic routing should apply and optionally, an indication of traffic correlation and/or an indication of application relocation possibility and/or indication of UE IP address preservation. The PCC rule also contains per DNAI a traffic steering policy ID and/or N6 traffic routing information, if the N6 traffic routing information is explicitly provided in the AF request.
If Traffic Correlation ID is included in the AF request, with EAS Correlation Indication or Common EAS and FQDN(s) parameters, the Traffic Correlation ID and the EAS Correlation Indication or Common EAS and FQDN(s) will be included in the PCC rule sent to SMF. The SMF can use the Traffic Correlation ID to determine that the UE belongs to a set of UEs identified by Traffic Correlation ID and a common EAS should be selected for the set of UE for the traffic identified by Traffic Descriptor as described in clause 6.2.3.2.5 of TS 23.548 [130].
If Traffic Correlation ID is included in the AF request, NEF updates the AF influence data in the UDR with the NEF Notification Endpoint to indicate it as responsible of the set of UEs associated with the Traffic correlation ID and to be notified with 5GC determined information as described in clause 6.2.3.2.7 of TS 23.548 [130].
The PCF provides in the PCC rule with information including the NEF Notification Endpoint for the SMF to notify to the NEF with 5GC determined information related to the UE members of the set of UEs identified by traffic correlation ID.
If Traffic Correlation ID and traffic correlation indication and FQDN(s) is included in the AF request, the Traffic Correlation ID and the traffic correlation indication will be included in the PCC rule sent to SMF. The SMF can use the Traffic Correlation ID to determine that the UE belongs to a set of UEs identified by Traffic Correlation ID and the UE needs to connect to EAS(s) corresponding to a common DNAI selected for the set of UE for the traffic identified by Traffic Descriptor, as described in clause 6.2.3.2.6 of TS 23.548 [130].
The PCF may also provide in the PCC rule information to subscribe the AF (or the NEF) to SMF events (UP path changes) corresponding to the AF request in which case it provides the information on AF subscription to corresponding SMF events received in the AF request. This is done by providing policies at PDU Session set-up or by initiating a PDU Session Modification procedure. When initiating a PDU Session set-up or PDU Session Modification procedure, the PCF considers the latest known UE location to determine the PCC rules provided to the SMF. The PCF evaluates the temporal validity condition of the AF request and informs the SMF to activate or deactivate the corresponding PCC rules according to the evaluation result. When policies specific to the PDU Session and policies general to multiple PDU Sessions exist, the PCF gives precedence to the PDU Session specific policies over the general policies. The PCF authorizes the AF request of User Plane Latency Requirements. If the PCF determines that the requirements can't be authorized, the PCF rejects the AF request.
The spatial validity condition is resolved at the PCF. In order to do that, the PCF subscribes to the SMF to receive notifications about change of UE location in an area of interest (i.e. Presence Reporting Area). The subscribed area of interest may be the same as spatial validity condition, or may be a subset of the spatial validity condition (e.g. a list of TAs) based on the latest known UE location. When the SMF detects that UE entered the area of interest subscribed by the PCF, the SMF notifies the PCF and the PCF provides to the SMF the PCC rules described above by triggering a PDU Session Modification. When the SMF becomes aware that the UE left the area subscribed by the PCF, the SMF notifies the PCF and the PCF provides updated PCC rules by triggering a PDU Session Modification. SMF notifications to the PCF about UE location in or out of the subscribed area of interest are triggered by UE location change notifications received from the AMF or by UE location information received during a Service Request or Handover procedure.
When the PCC rules are activated, the SMF may, based on local policies, take the information in the PCC rules and, optionally, the Service Experience analytics and/or DN Performance analytics per UP path (including UPF and/or DNAI and/or AS instance) as defined in clause 6.4.3 and clause 6.14.3, respectively, of TS 23.288 [86] into account to:
- (re)select UP paths (including DNAI(s)) for PDU Sessions. The SMF is responsible for handling the mapping between the UE location (TAI / Cell-Id) and DNAI(s) associated with UPF and applications and the selection of the UPF(s) that serve a PDU Session. This is described in clause 6.3.3. If the PDU Session is of IP type and if Indication of UE IP address preservation is included in the PCC rules, the SMF should preserve the UE IP address, by not reselecting the related PSA UPF once the PSA UPF is selected, for the traffic identified in the PCC rule. If the user plane latency requirement is included in the PCC rules, the SMF chooses the PSA UPF that satisfies the user plane latency requirement. If the PCC rules are related to a 5G VN group served by the SMF and if the PCC rule includes an indication of traffic correlation, the SMF should select a common DNAI for the PDU Sessions of the 5G VN group, see clause 5.29.
- configure traffic steering at UPF, including activating mechanisms for traffic multi-homing or enforcement of an UL Classifier (UL CL). Such mechanisms are defined in clause 5.6.4. This may include that the SMF is providing the UPF with packet handling instructions (i.e. PDRs and FARs) for steering traffic to the local access to the DN. The packet handling instructions are generated by the SMF using the traffic steering policy ID and/or the N6 traffic routing information in the PCC rules corresponding to the applied DNAI. In the case of UP path reselection, the SMF may configure the source UPF to forward traffic to the UL CL/BP so that the traffic is steered towards the target UPF.
- if Information on AF subscription to corresponding SMF events has been provided in the PCC rule, inform the AF of the (re)selection of the UP path (UP path change). If the information includes an indication of "AF acknowledgment to be expected", the SMF may decide to wait for a response from the AF before it activates the new UP path, as described in clause 5.6.7.2.
When an I-SMF is inserted for a PDU Session and if Local Offloading Management is not applied, the I-SMF insertion, relocation or removal to a PDU session shall be transparent (i.e. not aware) to the PCF and to the AF. When an I-SMF is inserted for a PDU Session and if Local Offloading Management is applied (see TS 23.548 [130]), the I-SMF insertion shall be transparent (i.e. not aware) to the AF but the insertion of I-SMF is not transparent to the PCF. The processing of the AF influence on traffic routing is described in clause 5.34 and detailed procedure is described in clause 4.23.6 of TS 23.502 [3]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.7.2 Enhancement of UP path management based on the coordination with AFs | In order to avoid or minimize service interruption during PSA relocation for a PDU session of SSC mode 3, or a PDU session with UL CL or branch point, according to the indication of "AF acknowledgment to be expected" on AF subscription to corresponding SMF events (DNAI change) (that may be provided in PCC rules received from the PCF as defined in clause 5.6.7.1 except in HR-SBO case or that may be provided directly by V-NEF to V-SMF as defined in TS 23.548 [130] clause 6.7.2 in case of HR-SBO) or according to local configuration (e.g. DN-related policies) the SMF may wait for a response from the AF after sending a notification (an early notification or a late notification) to the AF. In the case of late notification, based on the indication of "AF acknowledgment to be expected" on AF subscription, the SMF may send the notification before activating the UP path towards a new DNAI (possibly through a new PSA).
NOTE 1: Before the UP path toward the new DNAI is activated, application traffic data (if any exists) is still routed toward the old DNAI.
The notification sent from the SMF to the AF indicates UP path management events (DNAI change) as described in clause 5.6.7.1. The AF can confirm the DNAI change indicated in the notification with the SMF by sending a positive response to the notification to the SMF or reject the DNAI change by sending a negative response.
NOTE 2: The AF can determine whether application relocation is needed according to the notification of DNAI change. If not, the AF can send a positive response to the SMF immediately; otherwise, the AF sends the positive response after application relocation is completed or a negative response if the AF determines that the application relocation cannot be completed on time (e.g. due to temporary congestion). The AF decision and behaviours on application relocation are not defined. However, the new DNAI may be associated with a new AF. In such cases, the SMF and the old AF cancel earlier subscribed UP path management event notifications and the new AF subscribes to receive UP path management event notifications from the SMF.
The AF can include N6 traffic routing information related to the target DNAI in a positive response sent to the SMF. The SMF configures the N6 traffic routing information from the AF response to the PSA on the UP path.
The AF can include the EAS relocation Indication to indicate the application(s) to be relocated.
In the case of early notification, based on the indication of "AF acknowledgment to be expected" on AF subscription, the SMF does not configure the UP path towards the new DNAI until it receives a positive AF response as specified in clause 4.3.6.3 of TS 23.502 [3].
In the case of late notification, based on the indication of "AF acknowledgment to be expected" on AF subscription, the SMF does not activate the UP path towards the new DNAI until it receives a positive AF response as specified in clause 4.3.5 of TS 23.502 [3].
NOTE 3: After the UP path toward the new DNAI is activated, data is routed toward the new DNAI.
If the SMF receives a negative response at any time, the SMF keeps using the original DNAI and may cancel related PSA relocation or addition. The SMF may perform DNAI reselection afterwards if needed.
The SMF can assume according to local policy a negative response if a response is expected and but not received from the AF within a certain time window.
When Early/Late Notification happens, the SMF notifies AF about the target DNAI and may indicate capability of supporting EAS IP replacement in 5GC.When EAS relocation is performed, the AF sends an/a early/late notification response to the SMF after the EAS relocation is completed, which may include the Information for EAS IP Replacement in 5GC. The SMF may instruct the local PSA to use the FAR that contains Information elements of "IP Address and Port Number Replacement" as described in clause 6.3.3 of TS 23.548 [130]. If local PSA relocation is required, the SMF may request the target local PSA to buffer uplink traffic as described in clause 6.3.5 of TS 23.548 [130].
AF relocation may be triggered by SMF e.g. in relationship with DNAI change due to UE mobility. In the case of AF relocation involving different DNAI(s), it is possible that the source EHE is unaware of other/target EHE specific deployment details. In such cases, when SMF selects a target DNAI (e.g. based on current UE location), the SMF may determine based on the EDI that the target DNAI is not supported by the source AF. The SMF determines the target AF ID based on the target DNAI and the EDI. Accordingly, as part of Early/Late Notification, the SMF provides the target AF ID to the source AF as described in clause 4.3.6.3 of TS 23.502 [3]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.8 Selective activation and deactivation of UP connection of existing PDU Session | This clause applies to the case when a UE has established multiple PDU Sessions. The activation of a UP connection of an existing PDU Session causes the activation of its UE-CN User Plane connection (i.e. data radio bearer and N3 tunnel).
For the activation of a UP connection the service area restrictions as specified in clause 5.3.4.1.1 apply.
For the UE in the CM-IDLE state in 3GPP access, either UE or Network-Triggered Service Request procedure may support independent activation of UP connection of existing PDU Session. For the UE in the CM-IDLE state in non-3GPP access, UE-Triggered Service Request procedure allows the re-activation of UP connection of existing PDU Sessions and may support independent activation of UP connection of existing PDU Session.
A UE in the CM-CONNECTED state invokes a Service Request (see clause 4.2.3.2 of TS 23.502 [3]) procedure to request the independent activation of the UP connection of existing PDU Sessions.
Network Triggered re-activation of UP connection of existing PDU Sessions is handled as follows:
- If the UE CM state in the AMF is already CM-CONNECTED on the access (3GPP, non-3GPP) associated to the PDU Session in the SMF, the network may re-activate the UP connection of a PDU Session using a Network Initiated Service Request procedure.
Otherwise:
- If the UE is registered in both 3GPP and non-3GPP accesses and the UE CM state in the AMF is CM-IDLE in non-3GPP access, the UE can be paged or notified through the 3GPP access for a PDU Session associated in the SMF (i.e. last routed) to the non-3GPP access:
- If the UE CM state in the AMF is CM-IDLE in 3GPP access, the paging message may include the access type associated with the PDU Session in the SMF. The UE, upon reception of the paging message containing an access type, shall reply to the 5GC via the 3GPP access using the NAS Service Request message, which shall contain the list of PDU Sessions associated with the received access type and whose UP connections can be re-activated over 3GPP (i.e. the list does not contain the PDU Sessions whose UP connections cannot be re-activated on 3GPP based on UE policies and whether the S-NSSAIs of these PDU Sessions are within the Allowed NSSAI for 3GPP access). If the PDU Session for which the UE has been paged is in the list of the PDU Sessions provided in the NAS Service Request and the paging was triggered by pending DL data, the 5GC shall re-activate the PDU Session UP connection over 3GPP access. If the paging was triggered by pending DL signalling, the Service Request succeeds without re-activating the PDU session UP connection over the 3GPP access and the pending DL signalling is delivered to the UE over the 3GPP access;
- If the UE CM state in the AMF is CM-CONNECTED in 3GPP access, the notification message shall include the non-3GPP Access Type. The UE, upon reception of the notification message, shall reply to the 5GC via the 3GPP access using the NAS Service Request message, which shall contain the List of Allowed PDU Sessions that can be re-activated over 3GPP or an empty List of Allowed PDU Sessions if no PDU Sessions are allowed to be re-activated over 3GPP access.
NOTE: A UE that is in a coverage of a non-3GPP access and has PDU Session(s) that are associated in the UE (i.e. last routed) to non-3GPP access, is assumed to attempt to connect to it without the need to be paged.
- If the UE is registered in both 3GPP and non-3GPP accesses served by the same AMF and the UE CM state in the AMF is CM-IDLE in 3GPP access and is in CM-CONNECTED in non 3GPP access, the UE can be notified through the non-3GPP for a PDU Session associated in the SMF (i.e. last routed) to the 3GPP access. The notification message shall include the 3GPP Access Type. Upon reception of the notification message, when 3GPP access is available, the UE shall reply to the 5GC via the 3GPP access using the NAS Service Request message.
In addition to the above, a PDU Session may be established as an always-on PDU Session as described in clause 5.6.13.
The deactivation of the UP connection of an existing PDU Session causes the corresponding data radio bearer and N3 tunnel to be deactivated. The UP connection of different PDU Sessions can be deactivated independently when a UE is in CM-CONNECTED state in 3GPP access or non-3GPP access. At the deactivation of the UP of a PDU Session using a N9 tunnel whose end-point is controlled by an I-SMF, the N9 tunnel is preserved. If a PDU Session is an always-on PDU Session, the SMF should not deactivate a UP connection of this PDU Session due to inactivity. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9 Session and Service Continuity | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.1 General | The support for session and service continuity in 5G System architecture enables to address the various continuity requirements of different applications/services for the UE. The 5G System supports different session and service continuity (SSC) modes defined in this clause. The SSC mode associated with a PDU Session does not change during the lifetime of a PDU Session. The following three modes are specified with further details provided in the next clause:
- With SSC mode 1, the network preserves the connectivity service provided to the UE. For the case of PDU Session of IPv4 or IPv6 or IPv4v6 type, the IP address is preserved.
- With SSC mode 2, the network may release the connectivity service delivered to the UE and release the corresponding PDU Session(s). For the case of IPv4 or IPv6 or IPv4v6 type, the release of the PDU Session induces the release of IP address(es) that had been allocated to the UE.
- With SSC mode 3, changes to the user plane can be visible to the UE, while the network ensures that the UE suffers no loss of connectivity. A connection through new PDU Session Anchor point is established before the previous connection is terminated in order to allow for better service continuity. For the case of IPv4 or IPv6 or IPv4v6 type, the IP address is not preserved in this mode when the PDU Session Anchor changes.
NOTE: In this Release of the specification, the addition/removal procedure of additional PDU Session Anchor in a PDU Session for local access to a DN is independent from the SSC mode of the PDU Session. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.2 SSC mode | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.2.1 SSC Mode 1 | For a PDU Session of SSC mode 1, the UPF acting as PDU Session Anchor at the establishment of the PDU Session is maintained regardless of the access technology (e.g. Access Type and cells) a UE is successively using to access the network.
In the case of a PDU Session of IPv4 or IPv6 or IPv4v6 type, IP continuity is supported regardless of UE mobility events.
In this Release of the specification, when IPv6 multihoming or UL CL applies to a PDU Session of in SSC mode 1 and the network allocates (based on local policies) additional PDU Session Anchors to such a PDU Session, these additional PDU Session Anchors may be released or allocated and the UE does not expect that the additional IPv6 prefix is maintained during the lifetime of PDU Session.
SSC mode 1 may apply to any PDU Session type and to any access type.
A UE supporting PDU Connectivity shall support SSC mode 1. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.2.2 SSC Mode 2 | If a PDU Session of SSC mode 2 has a single PDU Session Anchor, the network may trigger the release of the PDU Session and instruct the UE to establish a new PDU Session to the same data network immediately. The trigger condition depends on operator policy e.g. request from Application Function, based on load status, etc. At establishment of the new PDU Session, a new UPF acting as PDU Session Anchor can be selected.
Otherwise, if a PDU Session of SSC mode 2 has multiple PDU Session Anchors (i.e. in the case of multi-homed PDU Sessions or in the case that UL CL applies to a PDU Session of SSC mode 2), the additional PDU Session Anchors may be released or allocated.
SSC mode 2 may apply to any PDU Session type and to any access type.
SSC mode 2 is optional to be supported in the UE.
NOTE 1: Features depending on SSC mode 2 will not work with the lack of support for SSC mode 2 in the UE.
NOTE 2: In UL CL mode, the UE is not involved in PDU Session Anchor re-allocation, so that the existence of multiple PDU Session Anchors is not visible to the UE. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.2.3 SSC Mode 3 | For PDU Session of SSC mode 3, the network allows the establishment of UE connectivity via a new PDU Session Anchor to the same data network before connectivity between the UE and the previous PDU Session Anchor is released. When trigger conditions apply, the network decides whether to select a PDU Session Anchor UPF suitable for the UE's new conditions (e.g. point of attachment to the network).
In this Release of specification, SSC mode 3 only applies to IP PDU Session type and to any access type.
In the case of a PDU Session of IPv4 or IPv6 or IPv4v6 type, during the procedure of change of PDU Session Anchor, the following applies:
a. For a PDU Session of IPv6 type, the new IP prefix anchored on the new PDU Session Anchor may be allocated within the same PDU Session (relying on IPv6 multi-homing specified in clause 5.6.4.3), or
b. The new IP address and/or IP prefix may be allocated within a new PDU Session that the UE is triggered to establish.
After the new IP address/prefix has been allocated, the old IP address/prefix is maintained during some time indicated to the UE via NAS signalling (as described in clause 4.3.5.2 of TS 23.502 [3]) or via Router Advertisement (as described in clause 4.3.5.3 of TS 23.502 [3]) and then released.
If a PDU Session of SSC mode 3 has multiple PDU Session Anchors (i.e. in the case of multi-homed PDU Sessions or in the case that UL CL applies to a PDU Session of SSC mode 3), the additional PDU Session Anchors may be released or allocated.
SSC mode 3 is optional to be supported in the UE.
NOTE: Features depending on SSC mode 3 will not work with the lack of support for SSC mode 3 in the UE. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.9.3 SSC mode selection | SSC mode selection is done by the SMF based on the allowed SSC modes -including the default SSC mode) in the user subscription as well as the PDU Session type and if present, the SSC mode requested by the UE.
The operator may provision a SSC mode selection policy (SSCMSP) to the UE as part of the URSP rule -see clause 6.6.2 of TS 23.503 [45]). The UE shall use the SSCMSP to determine the type of session and service continuity mode associated with an application or group of applications for the UE as described in clause 6.6.2.3 of TS 23.503 [45]. If the UE does not have SSCMSP, the UE can select a SSC mode based on UE Local Configuration as described in TS 23.503 [45], if applicable. If the UE cannot select a SSC mode, the UE requests the PDU Session without providing the SSC mode.
NOTE 1: The UE can use the SSC Mode Selection component of the URSP rule with match-all traffic descriptor if there is no SSC mode in the UE local configuration.
The SSC mode selection policy rules provided to the UE can be updated by the operator by updating the URSP rule.
The SMF receives from the UDM the list of allowed SSC modes and the default SSC mode per DNN per S-NSSAI as part of the subscription information.
If a UE provides an SSC mode when requesting a new PDU Session, the SMF selects the SSC mode by either accepting the requested SSC mode or rejecting the PDU Session Establishment Request message with the cause value and the SSC mode(s) allowed to be used back to UE based on the PDU Session type, subscription and/or local configuration. Based on that cause value and the SSC mode(s) allowed to be used, the UE may re-attempt to request the establishment of that PDU Session with the SSC mode allowed to be used or using another URSP rule.
If a UE does not provide an SSC mode when requesting a new PDU Session, then the SMF selects the default SSC mode for the data network listed in the subscription or applies local configuration to select the SSC mode.
SSC mode 1 shall be assigned to the PDU Session when static IP address/prefix is allocated to the PDU Session based on the static IP address/prefix subscription for the DNN and S-NSSAI. The SMF shall inform the UE of the selected SSC mode for a PDU Session.
The UE shall not request and the network shall not assign SSC mode 3 for the PDU Session of Unstructured type or Ethernet type.
NOTE 2: To avoid issues for UEs not supporting all SSC modes, the operator can, in the subscription data and local configuration, include at least SSC mode 1 in the allowed SSC modes and set default SSC mode to 1 (since all UEs supporting PDU sessions are mandated to support SSC mode 1). Still the 5GC can trigger PDU session release with a cause code indicating reactivation due to, e.g. restoration or user plane path optimization purposes, though this may cause interruption of the service. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.10 Specific aspects of different PDU Session types | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.10.1 Support of IP PDU Session type | The IP address allocation is defined in clause 5.8.1
The UE may acquire following configuration information from the SMF, during the lifetime of a PDU Session:
- Address(es) of P-CSCF(s);
- Address(es) of DNS server(s).
- If the UE indicates support of DNS with security as defined in TS 33.501 [29] to the network in PCO and the network wants to enforce the use of DNS with security, the configuration information sent by the SMF via PCO may also include the corresponding DNS server security information as specified in TS 24.501 [47] and TS 33.501 [29].
- the GPSI of the UE.
The UE may acquire from the SMF, at PDU Session Establishment, the MTU that the UE shall consider, see clause 5.6.10.4.
The UE may provide following information to the SMF during the lifetime of a PDU Session:
- an indication of the support of P-CSCF re-selection based on procedures specified in TS 24.229 [62] (clauses B.2.2.1C and L.2.2.1C).
- PS data off status of the UE.
NOTE 2: An operator can deploy NAT functionality in the network; a UPF supporting NAT information exposure functionality can register this capability with the NRF and the UPF can expose mapping between public and private IP addresses. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.10.2 Support of Ethernet PDU Session type | For a PDU Session set up with the Ethernet PDU Session type, the SMF and the UPF acting as PDU Session Anchor (PSA) can support specific behaviours related with the fact the PDU Session carries Ethernet frames.
Depending on operator configuration related with the DNN, different configurations for how Ethernet traffic is handled on N6 may apply, for example:
- Configurations with a 1-1 relationship between a PDU Session and a N6 interface possibly corresponding to a dedicated tunnel established over N6. In this case the UPF acting as PSA transparently forwards Ethernet frames between the PDU Session and its corresponding N6 interface and it does not need to be aware of MAC addresses used by the UE in order to route down-link traffic.
- Configurations, where more than one PDU Session to the same DNN (e.g. for more than one UE) corresponds to the same N6 interface. In this case the UPF acting as PSA needs to be aware of MAC addresses used by the UE in the PDU Session in order to map down-link Ethernet frames received over N6 to the appropriate PDU Session. Forwarding behaviour of the UPF acting as PSA is managed by SMF as specified in clause 5.8.2.5.
NOTE 1: The "MAC addresses used by the UE" correspond to any MAC address used by the UE or any device locally connected to the UE and using the PDU Session to communicate with the DN.
Based on operator configuration, the SMF may request the UPF acting as the PDU Session Anchor to respond to ARP/IPv6 Neighbour Solicitation requests based on local cache information, i.e. the mapping between the UE MAC address to the UE IP address and the DN where the PDU Session is connected to, or to redirect the ARP traffic from the UPF to the SMF. Responding to ARP/IPv6 ND based on local cache information applies to ARP/IPv6 ND received in both UL and DL directions.
NOTE 2: Responding to ARP/ND from a local cache assumes the UE or the devices behind the UE acquire their IP address via in-band mechanisms that the SMF/UPF can detect and by this link the IP address to the MAC address.
NOTE 3: This mechanism is intended to avoid broadcasting or multicasting the ARP/IPv6 ND to every UE.
Ethernet Preamble and Start of Frame delimiter are not sent over 5GS:
- For UL traffic the UE strips the preamble and frame check sequence (FCS) from the Ethernet frame.
- For DL traffic the PDU Session Anchor strips the preamble and frame check sequence (FCS) from the Ethernet frame.
Neither a MAC nor an IP address is allocated by the 5GC to the UE for a PDU Session.
The PSA shall store the MAC addresses received from the UE and associate those with the appropriate PDU Session.
The SMF may receive a list of allowed VLAN tags from DN-AAA (for a maximum of 16 VLAN tags) or as part of subscription data from UDM (for a maximum of 16 VLAN tags) or may be locally configured with allowed VLAN tags values.
The SMF may also determine instructions on VLAN handling (i.e. the C-TAG to be inserted UL and removed DL and/or S-TAG to be inserted UL and removed DL), based on local configuration, or VLAN handling information in subscription data from UDM, or VLAN handling information received from DN-AAA. The allowed VLAN tags and VLAN handling information received from DN-AAA takes precedence over the allowed VLAN tags and VLAN handling information in subscription data received from UDM. Similarly, the allowed VLAN tags and the VLAN handling information from UDM and DN-AAA takes precedence over the local configuration. Taking the list of allowed VLAN tags and instructions on VLAN handling into account, the SMF determines the VLAN handling for the PDU Session and instructs the UPF to accept or discard the UE traffic based on the allowed VLAN tags, as well as to handle VLAN tags (addition/removal) via PDR (Outer header removal) and FAR (UPF applying Outer header creation of a Forwarding policy). For example:
- The UPF may insert (for uplink traffic) and remove (for downlink traffic) a S-TAG on N6 or N19 or internal interface ("5G VN internal") for the traffic from and to the UE.
- The UPF may insert (for uplink traffic) and remove (for downlink traffic) a VLAN tag on the N6 interface while there is no VLAN in the traffic to and from the UE.
- The UPF may discard any UE traffic that does not contain any allowed VLAN tag when the UPF handles the UE uplink or downlink traffic.
NOTE 4: This can be used for traffic steering to N6-LAN but also for N6-based traffic forwarding related with 5G-VN service described in clause 5.29.4
NOTE 5: The VLAN tags in the list of allowed VLAN tags refers to the outer VLAN tag of Ethernet frame (which is the one closest to the Ethernet header) regardless of whether this outer VLAN tag corresponds to a S-TAG or to a C-TAG.
Apart from specific conditions related to the support of PDU sessions over W-5GAN defined in TS 23.316 [84], the UPF shall not remove VLAN tags sent by the UE and the UPF shall not insert VLAN tags for the traffic sent to the UE.
PDU(s) containing a VLAN tag shall be switched only within the same VLAN by a PDU Session Anchor.
The UE may acquire from the SMF, at PDU Session Establishment, the MTU of the Ethernet frames' payload that the UE shall consider, see clause 5.6.10.4.
NOTE 6: The UE may operate in bridge mode with regard to a LAN it is connecting to the 5GS, thus different MAC addresses may be used as source address of different frames sent UL over a single PDU Session (and destination MAC address of different frames sent DL over the same PDU Session).
NOTE 7: Entities on the LAN connected to the 5GS by the UE may have an IP address allocated by the DN but the IP layer is considered as an application layer which is not part of the Ethernet PDU Session.
NOTE 8: In this Release of the specification, only the UE connected to the 5GS is authenticated, not the devices behind such UE.
NOTE 9: 5GS does not support the scenario where a MAC address or if VLAN applies a (MAC address, VLAN) combination is used on more than one PDU Session for the same DNN and S-NSSAI.
NOTE 10: This Release of the specification does not guarantee that the Ethernet network remains loop-free. Deployments need to be verified on an individual basis that loops in the Ethernet network are avoided.
NOTE 11: This Release of the specification does not guarantee that the Ethernet network properly and quickly reacts to topology changes. Deployments need to be verified on an individual basis how they react to topology changes.
Different Frames exchanged on a PDU Session of Ethernet type may be served with different QoS over the 5GS. Thus, the SMF may provide to the UPF Ethernet Packet Filter Set and forwarding rule(s) based on the Ethernet frame structure and UE MAC address(es). The UPF detects and forwards Ethernet frames based on the Ethernet Packet Filter Set and forwarding rule(s) received from the SMF. This is further defined in clauses 5.7 and 5.8.2.
When a PDU Session of Ethernet PDU type is authorized by a DN as described in clause 5.6.6, the DN-AAA server may, as part of authorization data, provide the SMF with a list of allowed MAC addresses for this PDU Session; the list is limited to a maximum of 16 MAC addresses. When the list has been provided for a PDU Session, the SMF sets corresponding filtering rules in the UPF(s) acting as PDU Session Anchor for the PDU Session. The UPF discards any UL traffic that does not contain one of these MAC addresses as a source address if the list of allowed MAC addresses is provided.
In this Release of specification, the PDU Session of Ethernet PDU Session type is restricted to SSC mode 1 and SSC mode 2.
For a PDU Session established with the Ethernet PDU Session type, the SMF may, upon PCF request, need to ensure reporting to the PCF of all Ethernet MAC addresses used as UE address in a PDU Session. In this case, as defined in clause 5.8.2.12, the SMF controls the UPF to report the different MAC addresses used as source address of frames sent UL by the UE in the PDU Session.
NOTE 12: This relates to whether AF control on a per MAC address is allowed on the PDU Session as defined in clause 6.1.1.2 of TS 23.503 [45].
The PCF may activate or deactivate the reporting of the UE MAC address using the "UE MAC address change" Policy Control Request Trigger as defined in Table 6.1.3.5-1 of TS 23.503 [45].
The SMF may relocate the UPF acting as the PDU Session Anchor for an Ethernet PDU Session as defined in clause 4.3.5.8 of TS 23.502 [3]. The relocation may be triggered by a mobility event such as a handover, or may be triggered independent of UE mobility, e.g. due to load balancing reasons. In order to relocate the PSA UPF, the reporting of the UE MAC addresses needs to be activated by the SMF. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.10.3 Support of Unstructured PDU Session type | Different Point-to-Point (PtP) tunnelling techniques may be used to deliver Unstructured PDU Session type data to the destination (e.g. application server) in the Data Network via N6.
Point-to-point tunnelling based on UDP/IP encapsulation as described below may be used. Other techniques may be supported. Regardless of addressing scheme used from the UPF to the DN, the UPF shall be able to map the address used between the UPF and the DN to the PDU Session.
When Point-to-Point tunnelling based on UDP/IPv6 is used, the following considerations apply:
- IPv6 prefix allocation for PDU Sessions are performed locally by the (H-)SMF without involving the UE.
- The UPF(s) acts as a transparent forwarding node for the payload between the UE and the destination in the DN.
- For uplink, the UPF forwards the received Unstructured PDU Session type data to the destination in the data network over the N6 PtP tunnel using UDP/IPv6 encapsulation.
- For downlink, the destination in the data network sends the Unstructured PDU Session type data using UDP/IPv6 encapsulation with the IPv6 address of the PDU Session and the 3GPP defined UDP port for Unstructured PDU Session type data. The UPF acting as PDU Session Anchor decapsulates the received data (i.e. removes the UDP/IPv6 headers) and forwards the data identified by the IPv6 prefix of the PDU Session for delivery to the UE.
- The (H-)SMF performs the IPv6 related operations but the IPv6 prefix is not provided to the UE, i.e. Router Advertisements and DHCPv6 are not performed. The SMF assigns an IPv6 Interface Identifier for the PDU Session. The allocated IPv6 prefix identifies the PDU Session of the UE.
- For AF influence on traffic routing (described in clause 5.6.7), when the N6 PtP tunnelling is used over the DNAI and the AF provides, by value, information about N6 traffic routing requirements in the AF request, the AF provides N6 PtP tunnelling requirements (IPv6 address and UDP port of the tunnel end in the DN) as the N6 traffic routing information associated to the DNAI; when the SMF notifies the AF of UP path management events, it includes the N6 PtP tunnel information related to the UP (the IPv6 address and the 3GPP defined UDP port of the tunnel end at the UPF) as N6 traffic routing information in the notification.
In this Release of the specification there is support for maximum one 5G QoS Flow per PDU Session of Type Unstructured.
In this Release of specification, the PDU Session of Unstructured PDU Session type is restricted to SSC mode 1 and SSC mode 2.
The UE may acquire from the SMF, at PDU Session Establishment, the MTU that the UE shall consider, see clause 5.6.10.4. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.10.4 Maximum Transfer Unit size considerations | In order to avoid data packet fragmentation between the UE and the UPF acting as PSA, the link MTU size in the UE should be set to the value provided by the network as part of the IP configuration. The link MTU size for IPv4 is sent to the UE by including it in the PCO (see TS 24.501 [47]). The link MTU size for IPv6 is sent to the UE by including it in the IPv6 Router Advertisement message (see RFC 4861 [54]).
NOTE 1: Ideally the network configuration ensures that for PDU Session type IPv4v6 the link MTU values provided to the UE via PCO and in the IPv6 Router Advertisement message are the same. In cases where this condition cannot be met, the MTU size selected by the UE is unspecified.
When using a PDU Session type Unstructured, the maximum uplink packet size and when using Ethernet, the Ethernet frames' payload, that the UE should use may be provided by the network as a part of the session management configuration by encoding it within the PCO (see TS 24.501 [47]).
When using a PDU Session type Unstructured, to provide a consistent environment for application developers, the network shall use a maximum packet size of at least 128 octets (this applies to both uplink and downlink).
When the MT and the TE are separated, the TE may either be pre-configured to use a specific default MTU size or the TE may use an MTU size provided by the network via the MT. Thus, it is not always possible to set the MTU value by means of information provided by the network.
NOTE 2: In network deployments that have MTU size of 1500 octets in the transport network, providing a link MTU value of 1358 octets (as shown in Figure J-1) to the UE as part of the IP configuration information from the network will prevent the IP layer fragmentation within the transport network between the UE and the UPF. For network deployments that uniformly support transport with larger MTU size than 1500 octets (for example with ethernet jumbo frames of MTU size up to 9216 octets), providing a link MTU value of MTU minus 142 octets to the UE as part of the IP configuration information from the network will prevent the IP layer fragmentation within the transport network between the UE and the UPF. Link MTU considerations are discussed further in Annex J.
NOTE 3: As the link MTU value is provided as a part of the session management configuration information, a link MTU value can be provided during each PDU Session establishment. In this release, dynamic adjustment of link MTU for scenarios where MTU is not uniform across transport are not addressed. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.11 UE presence in Area of Interest reporting usage by SMF | When a PDU Session is established or modified, or when the user plane path has been changed (e.g. UPF re-allocation/addition/removal), SMF may determine an Area of Interest, e.g. based on UPF Service Area, subscription by PCF for reporting UE presence in Presence Reporting Area, etc.
For 3GPP access, the Area of Interest corresponds:
- either to Presence Information that may correspond to:
- a list of Tracking Areas; or
- a list of Presence Reporting Area ID(s) and optionally the elements comprising TAs and/or NG-RAN nodes and/or cells identifiers corresponding to the PRA ID(s); or
- a LADN DNN; or
- a LADN DNN and a S-NSSAI; or
- a S-NSSAI.
For Non-3GPP access, the Area of Interest corresponds to:
- N3GPP TAI (see clause 5.3.2.3).
For UE location change into or out of an "area of interest", the SMF subscribes to "UE mobility event notification" service provided by AMF for reporting of UE presence in Area of Interest as described in clause 5.3.4.4. The AMF may send the UE location to the SMF along with the notification, e.g. for UPF selection. Upon reception of a notification from AMF, the SMF determines how to deal with the PDU Session, e.g. reallocate UPF.
In the case of LADN, the SMF provides the LADN DNN to the AMF to subscribe to "UE mobility event notification" for reporting UE presence in LADN service area. Upon reception of a notification from the AMF, the SMF determines how to deal with the PDU Session as described in clause 5.6.5.
In the case of Partial Network Slice Support and Support for Network Slices with Network Slice Area of Service not matching deployed Tracking Areas as described in clauses 5.15.17 and 5.15.18, the SMF provides the S-NSSAI to the AMF to "UE mobility event notification" for reporting UE presence in slice restriction area. Upon reception of a notification from the AMF, the SMF determines how to deal with the PDU Session as described in clauses 5.15.17 and 5.15.18.
For use cases related to policy control and charging decisions, the PCF may subscribe to event reporting from the SMF or the AMF, for UE presence in a Presence Reporting Area.
A Presence Reporting Area can be:
- A "UE-dedicated Presence Reporting Area", defined in the subscriber profile and composed of a short list of TAs and/or NG-RAN nodes and/or cells identifiers in a PLMN; or derived from the Area of Interest provided by the Application Function to the PCF (see clause 5.6.7) and composed of a short list of TAs and/or NG-RAN nodes and/or cells identifiers in a PLMN; or
- A "Core Network predefined Presence Reporting Area", predefined in the AMF and composed of a short list of TAs and/or NG-RAN nodes and/or cells identifiers in a PLMN.
In the case of Change of UE Presence in Presence Reporting Area, for core network predefined Presence Reporting Area, the AMF determines the "area of interest" corresponding to the Presence Reporting Area Identifier(s), provided by the PCF or the SMF, as a list of TAIs and/or cell identifiers and/or NG-RAN node identifiers based on local configuration. For UE-dedicated Presence Reporting Areas, the subscription for UE location change notification for an "area of interest" shall contain the PRA Identifier(s) and the list(s) of TAs, or NG-RAN Node identifier and/or cell identifiers composing the Presence Reporting Area(s). For Core Network predefined Presence Reporting Areas, the subscription for UE location change notification for an "area of interest" shall contain the PRA identifier(s).
NOTE 1: If the Presence Reporting Area (PRA) and RAN Notification Area (RNA) are partially overlapping, the PCF will not get notified for the change of PRA when UE enters or leaves the PRA but remains in the RNA in CM-CONNECTED with RRC_INACTIVE state, because AMF is not informed.
Each Core Network predefined Presence Reporting Area can be configured with a priority level in the AMF. In order to prevent overload, the AMF may set the reporting for one or more of the received Presence Reporting Area(s) to inactive under consideration of the priority configured for each of Core Network predefined Presence Reporting Area(s), while storing the reporting request for this Presence Reporting Area in the UE context.
NOTE 2: Change of UE presence in Presence Reporting Area reporting does not apply to home routed roaming.
The AMF may be configured with a PRA identifier which refers to a Set of Core Network predefined Presence Reporting Areas. If the PCF subscribes to change of UE location for an area of interest for a Set of Presence reporting areas and provides a PRA identifier then the SMF may subscribe for event reporting for this Set of Presence Reporting Areas by only indicating this PRA Identifier in the area of interest. When the Presence Reporting Area(s) to be reported belong to a set of Core Network predefined Presence Reporting Areas in which the AMF is requested to report on change of UE presence, the AMF shall additionally add to the report the PRA Identifier of the Set of Core Network predefined Presence Reporting Areas.
Upon change of AMF, the PRA identifier(s) and if provided, the list(s) of Presence Reporting Area elements are transferred for all PDU sessions as part of MM Context information to the target AMF during the mobility procedure. If one or more Presence Reporting Area(s) was set to inactive, the target AMF may decide to reactivate one or more of the inactive Presence Reporting Area(s). The target AMF indicates per PDU session to the corresponding SMF/PCF the PRA identifier(s) and whether the UE is inside or outside the Presence Reporting Area(s) as well as the inactive Presence Reporting Area(s), if any.
NOTE 3: The target AMF cannot set the Presence Reporting Area(s) received from the source serving node to inactive.
The subscription may be maintained during the life of PDU Session, regardless of the UP activation state of PDU Session (i.e. whether UP connection of the PDU Session is activated or not).
SMF may determine a new area of interest and send a new subscription to the AMF with the new area of interest.
SMF un-subscribes to "UE mobility event notification" service when PDU Session is released. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.12 Use of Network Instance | The SMF may provide a Network Instance to the UPF in FAR and/or PDR via N4 Session Establishment or N4 Modification procedures.
NOTE 1: a Network Instance can be defined e.g. to separate IP domains, e.g. when a UPF is connected to 5G-ANs in different IP domains, overlapping UE IP addresses assigned by multiple Data Networks, transport network isolation in the same PLMN, etc.
NOTE 2: As the SMF can provide over N2 the Network Instance it has selected for the N3 CN Tunnel Info, the 5G AN does not need to provide Network Instance to the 5GC.
The SMF determines the Network Instance based on local configuration.
The SMF may determine the Network Instance for N3 and N9 interfaces, taking into account e.g. UE location, registered PLMN ID of UE, S-NSSAI of the PDU Session.
The SMF may determine the Network Instance for N6 interface taking into account e.g. (DNN, S-NSSAI) of the PDU Session.
The SMF may determine the Network Instance for N19 interface taking into account e.g. the (DNN, S-NSSAI) identifying a 5G VN group.
NOTE 3: As an example, the UPF can use the Network Instance included in the FAR, together with other information such as Outer header creation (IP address part) and Destination interface in the FAR, to determine the interface in UPF (e.g. VPN or Layer 2 technology) for forwarding of the traffic. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.13 Always-on PDU session | An always-on PDU Session is a PDU Session for which User Plane resources have to be activated during every transition from CM-IDLE mode to CM-CONNECTED state.
Based on an indication from upper layers, a UE may request to establish a PDU Session as an always-on PDU Session. The SMF decides whether the PDU Session can be established as an always-on PDU Session. In Home Routed roaming case, based on local policies, the V-SMF shall be involved to determine whether the PDU Session can be established as an always-on PDU Session.
If the UE requests the 5GC to modify a PDU Session, which was established in EPS, to an always-on PDU Session after the first inter-system change from EPS to 5GS, the SMF decides whether the PDU Session can be established as an always-on PDU Session based on the procedure described above.
The UE shall request activation of User Plane resources for always-on PDU Sessions even if there are no pending uplink data for this PDU Session or when the Service Request is triggered for signalling only or when the Service Request is triggered for paging response only.
If the UE has one or more established PDU Sessions which are not accepted by the network as always-on PDU Sessions and the UE has no uplink user data pending to be sent for those PDU Sessions, the UE shall not request for activating User Plane resources for those PDU sessions. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.14 Support of Framed Routing | Framed Routing is only defined for PDU Sessions of the IP type (IPv4, IPv6, IPv4v6) and allows to support an IP network behind a UE, such that a range of IPv4 addresses or IPv6 prefixes is reachable over a single PDU Session, e.g. for enterprise connectivity. Framed Routes are IP routes behind the UE.
A PDU Session may be associated with multiple Framed Routes. Each Framed Route refers to a range of IPv4 addresses (i.e. an IPv4 address and an IPv4 address mask) or a range of IPv6 Prefixes (i.e. an IPv6 Prefix and an IPv6 Prefix length). The set of one or more Framed Routes associated to a PDU Session is contained in the Framed Route information. The network does not send Framed Route information to the UE: devices in the network(s) behind the UE get their IP address by mechanisms out of the scope of 3GPP specifications. See RFC 2865 [73], RFC 3162 [74].
Framed Route information is provided by the SMF to the UPF (acting as PSA) as part of Packet Detection Rule (PDR, see clause 5.8.5.3) related with the network side (N6) of the UPF.
NOTE: SMF can take the UPF capabilities into account when selecting PSA UPF, to ensure that the SMF chooses PSA UPF(s) that support Framed Routing for PDU Sessions to DNN and/or slices deemed to support Framed Routing e.g. DNN and/or slices intended to support RG or if Framed Route information has been received as part of Session Management Subscription data.
The Framed Route information may be provided to the SMF by:
- the DN-AAA server as part of PDU Session Establishment authentication/authorization by a DN-AAA server (as defined in clause 5.6.6); or by
- Session Management Subscription data associated with DNN and S-NSSAI sent by UDM (as defined in clause 5.2.3.3.1 of TS 23.502 [3]).
If the SMF receives Framed Route information both from DN-AAA and from UDM, the information received from DN-AAA takes precedence and supersedes the information received from UDM.
The IPv4 address / IPv6 Prefix allocated to the UE as part of the PDU Session establishment (e.g. delivered in NAS PDU Session Establishment Accept) may belong to one of the Framed Routes associated with the PDU Session or may be dynamically allocated outside of such Framed Routes.
If PCC applies to the PDU Session, at PDU Session establishment the SMF reports to the PCF the Framed Route information corresponding to the PDU Session (as described in clause 6.1.3.5 of TS 23.503 [45]). In this case, in order to support session binding, the PCF may further report to the BSF the Framed Route information corresponding to the PDU Session (as described in clause 6.1.2.2 of TS 23.503 [45]).
If the UDM or DN-AAA updates the Framed Route information during the lifetime of the PDU Session, the SMF releases the PDU Session and may include in the release request an indication for the UE to re-establish the PDU Session. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.15 Triggers for network analytics | Triggers for the SMF to request for or subscribe to the analytics information from the NWDAF are internal logic may include for example:
- UE PDU Session related event subscription by other NFs (e.g. AMF, NEF);
- UE access and mobility event reports from the AMF;
- locally detected events;
- analytics information received.
The trigger conditions may depend on operator and implementation policy in the SMF. When a trigger condition happens, the SMF may decide if any analytics information is needed and if so, request for or subscription to the analytics information from the NWDAF.
The SMF may, upon detection of certain local events, e.g. number of PDU sessions establishment or released reaches a threshold in a specific area, request for or subscribe to network analytics related to "Abnormal behaviour" as described in TS 23.288 [86] to detect whether there are any exceptional UE behaviours in this area. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.16 Support for Service Function Chaining | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.16.1 General | Service Function Chaining, also called N6-LAN Traffic Steering, refers to the steering of subscriber's traffic flows to appropriate operator or 3rd party Service Functions (e.g. NAT, antimalware, parental control, DDoS protection) in the N6-LAN.
The content of this clause applies to non-roaming and to Home Routed roaming scenario, i.e. to cases where the involved entities (AF, PCF, SMF, UPF) belong to the Home PLMN and the AF has an agreement with the Home PLMN.
The PCF controls Service Function Chaining by provisioning and modifying traffic steering control information for N6-LAN Traffic Steering as described in TS 23.503 [45], e.g. clause 6.1.3.14. The traffic steering control information for N6-LAN Traffic Steering consists of a traffic description and a reference to a traffic steering policy that is configured in the SMF/UPF.
The PCF derives the TSP ID(s) (that can be different for uplink and downlink directions) based on operator configuration and sends the TSP ID(s) to the SMF as part of the N6-LAN Traffic Steering Enforcement Control information in the PCC rule as described in clause 6.3.1 of TS 23.503 [45].
When the PCC rule is activated or updated with N6-LAN Traffic Steering Enforcement Control information, the SMF sets the Forwarding Policy (uplink and/or downlink) within the FAR(s) based on the authorized TSP ID(s) in the PCC rule and under consideration of the direction. In case that Application Function influence on traffic routing Enforcement Control information and N6-LAN Traffic Steering Enforcement Control information are both provided for the uplink direction, the SMF shall derive N4 rules which instruct the UPF to pass the traffic through the relevant Service Function(s) deployed in the N6-LAN before steering the traffic to the local data network. The SMF provides instructions to UPF for N6-LAN traffic steering as further detailed in clause 5.8.5.6.
The UPF applies traffic steering mechanism based on Forwarding Policy, i.e. the UPF performs deployment specific actions as configured for the Forwarding Policy.
NOTE 1: It is assumed that all UPFs in the operator network serving as PSA for the DNN/S-NSSAI/DNAI subject to N6-LAN traffic steering need to be configured with the same traffic steering information for N6-LAN traffic steering.
When performing deployment specific actions configured for the Forwarding Policy, the UPF may support traffic steering related functionality and user plane encapsulation protocols that are out of 3GPP scope (e.g. as defined by other standards organizations).
NOTE 2: The existing user plane mechanisms (e.g. VXLAN, NSH, GENEVE, GRE, VLAN, etc.) defined at IETF are reused as applicable by the PSA UPF to support N6-LAN traffic steering.
The mechanism used for forwarding the traffic between the Service Functions within the N6-LAN is out of 3GPP scope. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.16.2 Application Function influence on Service Function Chaining | An AF may request the steering of user plane traffic to a pre-configured chain of Service Functions on N6-LAN.
In the non-roaming scenario, Application Function influence on Service Function Chaining and Application Function influence on traffic routing (as defined in clause 5.6.7) can be applicable to the same traffic simultaneously.
It is assumed that a service level agreement exists between the operator and a third party that includes a list of authorized predefined Service Function Chains (SFCs), each SFC being identified based on a Service Function Chaining identifier (SFC ID). The AF may request the selected traffic flows to be steered towards a specific SFC, either at PDU Session establishment or any time after PDU Session establishment.
The AF requests may be sent to the PCF via the NEF. When SFC ID is included in the AF request, the parameters listed in Table 5.6.16.2-1 may be included in the AF request.
Table 5.6.16.2-1: Information element contained in AF request
Information Name
Applicable for PCF or NEF (NOTE 1)
Applicable for NEF only
Category
Traffic Description
Defines the target traffic to be influenced, represented by the combination of DNN and optionally S-NSSAI and application identifier or traffic filtering information.
The target traffic can be represented by AF-Service-Identifier, instead of combination of DNN and optionally S-NSSAI.
Mandatory
Target UE Identifier(s)
Indicates the UE(s) that the request is targeting, i.e. an individual UE, a group of UE represented by Internal Group Identifier (NOTE 2), or any UE accessing the combination of DNN and S-NSSAI.
GPSI can be applied to identify the individual UE, or External Group Identifier can be applied to identify a group of UE.
Mandatory
Spatial Validity Condition
Indicates that the request applies only to the traffic of UE(s) located in the specified location, represented by areas of validity.
The specified location can be represented by geographical area.
Optional
AF transaction identifier
The AF transaction identifier refers to the AF request.
N/A
Mandatory
SFC identifier(s)
Indicates the pre-defined Service Function Chain in downlink and/or uplink.
N/A
Mandatory
Metadata
Contains information that is transparently passed to UPF (NOTE 3) and provided by UPF to the Service Functions in N6-LAN.
N/A
Optional
NOTE 1: When the AF request targets existing or future PDU Sessions of multiple UE(s) or of any UE and is sent via the NEF, as described in clause 6.3.7.2, the information is stored in the UDR by the NEF and notified to the PCF by the UDR.
NOTE 2: Internal Group ID can only be used by an AF controlled by the operator and only towards PCF.
NOTE 3: The NEF, PCF and SMF do not need to understand the Metadata.
The PCF checks whether the SFC ID received from the AF corresponds to an authorized predefined SFC according to the service level agreement with this AF. Based on the SFC ID received from the AF, the PCF derives the TSP ID(s) (that can be different for uplink and downlink directions) and sends the TSP ID(s) and optionally Metadata (as provided by the AF) to the SMF as part of the PCC rule(s) as described in clause 6.3.1 of TS 23.503 [45].
The SMF behaves in the same way it is described in clause 5.6.16.1. If the SMF has received Metadata in the N6-LAN Traffic Steering Enforcement Control information of the PCC rule, the SMF forwards the Metadata to the UPF via N4 in the corresponding FAR as described in clause 5.8.5.6. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.17 Handling of Payload Headers | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.17.1 General | Handling of Payload Headers is an optional feature that allows exchange of information in-band in the user plane packets between the application (in the UE or in the application server) and UPF in the 5GS by supporting detection, insertion, replacement, or removal of payload headers of PDUs. Handling of Payload Headers is requested by the AF.
NOTE 1: The Handling of Payload Headers is not limited to the protocol header but can be any field or tag supported by a protocol.
NOTE 2: Handling of Payload Headers is expected to be used instead of Header Enrichment (see clause 5.8.5.6) as it allows more actions than just insertion, supports different protocols and allows the usage of actions in both directions.
The content of this clause applies to non-roaming and to Home Routed roaming scenario, i.e. to cases where the involved entities (AF, PCF, SMF, UPF) belong to the Home PLMN and the AF has an agreement with the Home PLMN.
The PCF controls Handling of Payload Headers in UPF by provisioning Header Handling Control information in the PCC rule. The PCF derives the Header Handling Control information as described in 6.1.3.30 of TS 23.503 [45] based on the AF request and provides it to the SMF in the related PCC rule as described in clause 6.3.1 of TS 23.503 [45]. SMF instructs UPF via N4 interactions by means of FAR (see clause 5.8.5.6).
The UPF performs Handling of Payload Headers and may notify events related to the Handling of Payload Headers to SMF, to NEF/AF or to both as described in clause 5.6.17.2. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.6.17.2 Application Function influence on Handling of Payload Headers | An AF may request the Handling of Payload Headers in order to detect, insert, replace, or remove payload headers of PDUs as well as to subscribe to notifications on events related to detection of payload headers of PDUs or related to the actions performed on payload headers of PDUs.
For the Handling of Payload Headers, it is assumed that a service level agreement (SLA) exists between the operator and the third party AF. The mechanism(s) required for detection and actions performed on payload headers (e.g. protocol layer, type of encryption, etc.) is agreed as part of the SLA. The reference(s) to the UPF configuration(s) corresponding to the header detection as well as reference(s) to SMF configuration corresponding to SMF context information are also part of the SLA. The AF is required to provide such reference(s) in the AF requests for Handling of Payload Headers.
In the non-roaming scenario, the features Handling of Payload Headers, Application Function influence on Service Function Chaining (as defined in clause 5.6.16) and Application Function influence on traffic routing (as defined in clause 5.6.7) can be applied simultaneously.
In the Home Routed roaming scenario, the features Handling of Payload Headers and Application Function influence on Service Function Chaining (as defined in clause 5.6.16) can be applied simultaneously.
The AF requests for Handling of Payload Headers are sent to the PCF via N5 (in the case of requests targeting specific on-going PDU Sessions of individual UE(s), for an AF allowed to interact directly with the 5GC NFs) or via the NEF. The AF requests that target existing or future PDU Sessions of multiple UE(s) or of any UE are sent via the NEF and may target multiple PCF(s), as described in clause 6.3.7.2. The PCF(s) transform(s) the Header Handling Control information provided in the AF requests into Header Handling Control information of the PCC rules that apply to the respective PDU Sessions. When the AF has subscribed to events related to Handling of Payload Headers, the enforcement of a requested action is a Header Handling event, and such event notifications are sent by the UPF(s) either directly to the AF or via an NEF.
When the request is sent via NEF, Nnef_TrafficInfluence service is used. The procedures for AF to request Handling of Payload Headers are described in clause 4.3.6 in TS 23.502 [3].
The AF requests may contain the information as described in the Table 5.6.17.2-1:
Table 5.6.17.2-1: Information elements contained in AF request
Information Name
Applicable for PCF or NEF (NOTE 1)
Applicable for NEF only
Category
Traffic Description
Defines the target traffic on which to apply handling of headers, represented by the combination of DNN and optionally S-NSSAI and application identifier or traffic filtering information.
The target traffic can be represented by AF-Service-Identifier.
Mandatory
Target UE Identifier(s)
Indicates the UE(s) that the request is targeting, i.e. an individual UE, a group of UE represented by Internal Group Identifier(s) (NOTE 2), or any UE accessing the combination of DNN and S-NSSAI.
GPSI can be applied to identify the individual UE, or External Group Identifier can be applied to identify a group of UEs.
Mandatory
Spatial Validity Condition
Indicates that the request applies only to the traffic of UE(s) located in the specified location, represented by areas of validity.
The specified location can be represented by geographical area.
Optional
Temporal Validity Condition
Time interval(s) when the request applies or duration(s).
N/A
Optional
AF transaction identifier
The AF transaction identifier refers to the AF request.
N/A
Mandatory
Header Handling Control information
Header Detection Reference
A reference to a UPF configuration which defines how to detect the protocol or the message in a protocol for which to perform the header handling actions.
N/A
Mandatory
Header Detection Support Information
(NOTE 3)
Any dynamic information provided by the AF which is required for the detection of the headers and cannot be preconfigured for the Header Detection Reference.
N/A
Optional
Header Handling Reporting Endpoint
Notification endpoint, i.e. per notification endpoint a Notification Target Address and a Notification Correlation ID.
N/A
Optional
Header Handling Control Reference
(NOTE 5)
A reference to a Header Handling Action related information pre-configured in the UPF.
N/A
Optional
Header Handling Direction
(NOTE 4)
Indicates if the header handling applies to UL or DL direction.
N/A
Optional
Header Handling Action
(NOTE 4)
Indicates the action to be performed on a specific header field.
N/A
Optional
Header Information
(NOTE 4)
A reference to a UPF configuration which defines how to identify or build a specific header field for which to perform the header handling action.
N/A
Optional
Header Value
(NOTE 4)
A string providing the value (or a reference to information to be provided by SMF) of the specific header field.
N/A
Optional
Header Handling Condition
(NOTE 4)
Indicates the condition for performing the header handling action.
N/A
Optional
Header Handling Reporting
(NOTE 4)
Indicates whether reporting is requested for the performed Header Handling Action.
N/A
Optional
NOTE 1: When the AF request targets existing or future PDU Sessions of multiple UE(s) or of any UE and is sent via the NEF, as described in clause 6.3.7.2, the information is stored in the UDR by the NEF and notified to the PCF by the UDR.
NOTE 2: Internal Group ID can only be used by an AF controlled by the operator and only towards PCF.
NOTE 3: The NEF, PCF and SMF do not need to understand the Header Detection Support Information.
NOTE 4: These parameters are provided together to request a Header Handling Action. Multiple sets of these parameters (and thus multiple header handling actions) can be provided.
NOTE 5: If a Header Handling Control Reference is provided, one or more of these parameters (marked by NOTE 4) can be provided to overwrite Header Handling Control information that is pre-configured in the UPF as per the Header Handling Control Reference. Multiple Header Handling Control References can be provided.
Traffic Description, Target UE identifier(s) and AF transaction identifier are common to all features requested simultaneously in the same AF request, more specifically, Application Function influence on Handling of Payload Headers, Application Function influence on Service Function Chaining (as defined in clause 5.6.16) and/or Application Function influence on traffic routing (as defined in clause 5.6.7). Spatial and Temporary validity conditions are optionally provided for a specific feature.
For each information element mentioned above as part of Header Handling Control Information in the AF request, a detailed description follows:
1) Header Detection Reference:
A reference to a UPF configuration which defines how to detect the protocol or the message in a protocol for which to perform the header handling actions (e.g. information about protocol layer, type of encryption, message type etc.).
2) Header Detection Support Information:
Header Detection Support Information is not standardised, but it can be interpreted by UPF based on SLA. It is sent transparently by NEF, PCF and SMF to UPF. It includes dynamic information provided by the AF which is required for the detection of the headers and cannot be preconfigured for the Header Detection Reference in UPF.
3) Header Handling Reporting Endpoints:
The notification endpoint provided by the AF for the notifications related to the Handling of Payload Headers. For the notification endpoint, a Notification Target Address and a Notification Correlation ID has to be provided.
4) Header Handling Direction:
Header Handling Direction indicates to which direction the Header Handling Action applies to. It can have the values uplink (UL) or downlink (DL).
5) Header Handling Action:
The action to be performed on a specific header field. One of the following can be requested:
- Detect. It is used to request the detection of a header field that is identified based on the UPF configuration referenced by the Header Information parameter and the information contained in the Header Value parameter, if it is provided.
- Remove. It is used to request the removal of a header field that is identified based on the UPF configuration referenced by the Header Information parameter and the information contained in the Header Value parameter, if it is provided.
- Replace. It is used to request the replacement of information in a header field that is identified based on the UPF configuration referenced by the Header Information parameter. The information to be replaced in the header field is also defined by that UPF configuration. The Header Value contains information that is used for the replacement.
- Insert. It is used to add a header field according to the UPF configuration referenced by the Header Information parameter. The Header Value contains information that is used for the insertion, if it is provided.
When multiple actions are requested on the target traffic, the enforcement of the Header Handling Action is performed in the following order: Detect, Remove, Replace and Insert.
NOTE 1: The execution order is relevant to build a request that has the intended impact on the traffic. For example, a header that is removed cannot be replaced.
6) Header Information:
A reference to a UPF configuration which defines how to identify or build a specific header field for which to perform the header handling action.
7) Header Value:
A string providing the value (or a reference to information to be provided by SMF) of the specific header field relevant for the action (i.e. to detect in case of actions Detect or Remove and for updating in case of actions Insert or Replace). Header Value may optionally be provided apart from the Replace action for which it is mandatory.
8) Header Handling Condition:
This defines how to apply the Header Handling Action. It may have one of the following values: every match or only for the first match.
NOTE 2: In addition, Spatial Validity Condition and Temporary Validity Condition parameters in the request can be used to indicate that the request applies only in certain UE location(s) or during certain time interval(s), respectively.
9) Header Handling Reporting:
Indicates whether reporting is requested for the performed Header Handling Action or not. In addition, to reduce signalling, the consumer may provide:
- Reporting suggestion information as described in clause 5.8.2.17.
- One-Time-Report indication, if the reporting of a first occurrence of the action per packet flow is enough.
When reporting is requested for an Action, the AF shall also provide Header Handling Reporting Endpoint.
10) Header Handling Control Reference:
A reference to a Header Handling Action related information that is pre-configured in the UPF. This pre-configuration in UPF comprise parameters 4-9. The Header Handling Control Reference can be provided in the AF request instead of providing the parameters of the Header Handling Control Information individually.
NOTE 3: The reference to a Header Handling Control information that is pre-configured in the UPF and the corresponding header handling behaviour has to be agreed as part of the SLA. In a deployment, for simplicity, an operator can choose to set this reference to the same value as the Header Handling Detection Reference. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7 QoS model | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1 General Overview | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.1 QoS Flow | The 5G QoS model is based on QoS Flows. The 5G QoS model supports both QoS Flows that require guaranteed flow bit rate (GBR QoS Flows) and QoS Flows that do not require guaranteed flow bit rate (Non-GBR QoS Flows). The 5G QoS model also supports Reflective QoS (see clause 5.7.5).
The QoS Flow is the finest granularity of QoS differentiation in the PDU Session. A QoS Flow ID (QFI) is used to identify a QoS Flow in the 5G System. User Plane traffic with the same QFI within a PDU Session receives the same traffic forwarding treatment (e.g. scheduling, admission threshold). The QFI is carried in an encapsulation header on N3 (and N9) i.e. without any changes to the e2e packet header. QFI shall be used for all PDU Session Types. The QFI shall be unique within a PDU Session. The QFI may be dynamically assigned or may be equal to the 5QI (see clause 5.7.2.1).
Within the 5GS, a QoS Flow is controlled by the SMF and may be preconfigured, or established via the PDU Session Establishment procedure (see clause 4.3.2 of TS 23.502 [3]), or the PDU Session Modification procedure (see clause 4.3.3 of TS 23.502 [3].
Any QoS Flow is characterised by:
- A QoS profile provided by the SMF to the AN via the AMF over the N2 reference point or preconfigured in the AN;
- One or more QoS rule(s) and optionally QoS Flow level QoS parameters (as specified in TS 24.501 [47]) associated with these QoS rule(s) which can be provided by the SMF to the UE via the AMF over the N1 reference point and/or derived by the UE by applying Reflective QoS control; and
- One or more UL and DL PDR(s) provided by the SMF to the UPF.
Within the 5GS, a QoS Flow associated with the default QoS rule is required to be established for a PDU Session and remains established throughout the lifetime of the PDU Session. This QoS Flow should be a Non-GBR QoS Flow (further details are described in clause 5.7.2.7).
A QoS Flow is associated with QoS requirements as specified by QoS parameters and QoS characteristics.
NOTE: The QoS Flow associated with the default QoS rule provides the UE with connectivity throughout the lifetime of the PDU Session. Possible interworking with EPS motivates the recommendation for this QoS Flow to be of type Non-GBR.
A QoS Flow may be enabled with PDU Set based QoS handling as described in clause 5.37.5. For such QoS Flows, PDU Set QoS Parameters (see clause 5.7.7) are determined by the PCF and provided by SMF to the 5G-AN as part of the QoS profile. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.2 QoS Profile | A QoS Flow may either be 'GBR' or 'Non-GBR' depending on its QoS profile. The QoS profile of a QoS Flow is sent to the (R)AN and it contains QoS parameters as described below (details of QoS parameters are described in clause 5.7.2):
- For each QoS Flow, the QoS profile shall include the QoS parameters:
- 5G QoS Identifier (5QI); and
- Allocation and Retention Priority (ARP).
- For each QoS Flow, the QoS profile may also include the QoS parameters:
- PDU Set QoS Parameters (described in clause 5.7.7).
- For each Non-GBR QoS Flow only, the QoS profile may also include the QoS parameter:
- Reflective QoS Attribute (RQA).
- For each GBR QoS Flow only, the QoS profile shall also include the QoS parameters:
- Guaranteed Flow Bit Rate (GFBR) - UL and DL; and
- Maximum Flow Bit Rate (MFBR) - UL and DL; and
- In the case of a GBR QoS Flow only, the QoS profile may also include one or more of the QoS parameters:
- Notification control;
- Maximum Packet Loss Rate - UL and DL.
NOTE: In this Release of the specification, the Maximum Packet Loss Rate (UL, DL) is only provided for a GBR QoS Flow belonging to voice media.
Each QoS profile has one corresponding QoS Flow identifier (QFI) which is not included in the QoS profile itself.
The usage of a dynamically assigned 5QI for a QoS Flow requires in addition the signalling of the complete 5G QoS characteristics (described in clause 5.7.3) as part of the QoS profile.
When a standardized or pre-configured 5QI is used for a QoS Flow, some of the 5G QoS characteristics may be signalled as part of the QoS profile (as described in clause 5.7.3).
5.7.1.2a Alternative QoS Profile
The Alternative QoS Profile(s) can be optionally provided for a GBR QoS Flow with Notification control enabled. If the corresponding PCC rule contains the related information (as described in TS 23.503 [45]), the SMF shall provide, in addition to the QoS profile, a prioritized list of Alternative QoS Profile(s) to the NG-RAN. If the SMF provides a new prioritized list of Alternative QoS Profile(s) to the NG-RAN (if the corresponding PCC rule information changes), the NG-RAN shall replace any previously stored list with it.
An Alternative QoS Profile represents a combination of QoS parameters PDB, PER, Averaging Window and GFBR to which the application traffic is able to adapt. The Alternative QoS Profile(s) shall also contain the PDU Set QoS parameters PSDB and PSER in the respective direction(s) when the QoS Profile contains PDU Set QoS parameters PSDB and PSER in UL and/or DL as described in clause 5.7.7; otherwise, the corresponding Alternative QoS Profile(s) shall not contain the PDU Set QoS parameters PSDB and PSER.
NOTE 1: There is no requirement that the GFBR monotonically decreases, nor that the PDB or PER (or PSDB or PSER) monotonically increase as the Alternative QoS Profiles become less preferred.
For a GBR QoS Flow using the Delay-critical resource type, an Alternative QoS Profile may also include the QoS parameter MDBV.
For the NG-RAN determination whether the GFBR, the PDB and the PER can be fulfilled (i.e. for all Notification control scenarios described in clause 5.7.2.4), the NG-RAN shall also apply the value of the optional QoS parameter present in an Alternative QoS Profile instead of the corresponding QoS parameter's value in the QoS Profile (which is either provided explicitly or implicitly via the standardized 5QI QoS characteristics defined in clause 5.7.4) for every optional QoS parameter present in an Alternative QoS Profile (e.g. MDBV).
NOTE 2: For the NG-RAN behaviour related to the mandatory QoS parameters in the Alternative QoS Profile (i.e. GFBR, PDB, PER), see clause 5.7.2.4.
When the NG-RAN sends a notification to the SMF that the QoS profile is not fulfilled, the NG-RAN shall, if the currently fulfilled values match an Alternative QoS Profile, include also the reference to the Alternative QoS Profile to indicate the QoS that the NG-RAN currently fulfils (see clause 5.7.2.4). The NG-RAN shall enable the SMF to determine when an NG-RAN node supports the Alternative QoS Profile feature but cannot fulfil even the least preferred Alternative QoS Profile.
NOTE 3: To reduce the risk that GBR QoS Flows are released in case of RAN resource limitations (and then experience difficulties in being re-established), Application Functions can set the least preferred Alternative Service Requirement to an undemanding level. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.3 Control of QoS Flows | The following options are supported to control QoS Flows:
1) For Non-GBR QoS Flows and when standardized 5QIs or pre-configured 5QIs are used and when the 5QI is within the range of the QFI (i.e. a value less than 64), the 5QI value may be used as the QFI of the QoS Flow.
(a) A default ARP shall be pre-configured in the AN; or
(b) The ARP and the QFI shall be sent to RAN over N2 at PDU Session Establishment or at PDU Session Modification and when NG-RAN is used every time the User Plane of the PDU Session is activated; and
2) For all other cases (including GBR and Non-GBR QoS Flows), a dynamically assigned QFI shall be used. The 5QI value may be a standardized, pre-configured or dynamically assigned. The QoS profile and the QFI of a QoS Flow shall be provided to the (R)AN over N2 at PDU Session Establishment/Modification and when NG-RAN is used every time the User Plane of the PDU Session is activated.
Only options 1b and 2 may apply to 3GPP ANs. Options 1a, 1b and 2 may apply to Non-3GPP access.
NOTE: Pre-configured 5QI values can be used when the UE is roaming if allowed by SLA between the operators. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.4 QoS Rules | The UE performs the classification and marking of UL User plane traffic, i.e. the association of UL traffic to QoS Flows, based on QoS rules. These QoS rules may be explicitly provided to the UE (i.e. explicitly signalled QoS rules using the PDU Session Establishment/Modification procedure), pre-configured in the UE or implicitly derived by the UE by applying Reflective QoS (see clause 5.7.5). A QoS rule contains the QFI of the associated QoS Flow, a Packet Filter Set (see clause 5.7.6) and a precedence value (see clause 5.7.1.9). An explicitly signalled QoS rule contains a QoS rule identifier which is unique within the PDU Session and is generated by SMF.
There can be more than one QoS rule associated with the same QoS Flow (i.e. with the same QFI).
When the UE informs the network about the number of supported Packet Filters for signalled QoS rules for the PDU Session (during the PDU Session Establishment procedure or using the PDU Session Modification procedure as described in clause 5.17.2.2.2 after the first inter-system change from EPS to 5GS for a PDU Session established in EPS and transferred from EPS with N26 interface), the SMF shall ensure that the sum of the Packet Filters used by all signalled QoS rules for a PDU Session does not exceed the number indicated by the UE.
A default QoS rule is required to be sent to the UE for every PDU Session establishment and it is associated with a QoS Flow. For IP type PDU Session or Ethernet type PDU Session, the default QoS rule is the only QoS rule of a PDU Session which may contain a Packet Filter Set that allows all UL packets and in this case, the highest precedence value shall be used for the QoS rule.
NOTE 2: How the UE evaluates UL packets against the Packet Filter Set in a QoS rule is described in clause 5.7.1.5.
NOTE 3: The QoS rule pre-configured in the UE is only used together with option 1a for control QoS Flows as described in clause 5.7.1.3. How to keep the consistency of QFI and Packet Filter Set between UE and network is out of scope in this release of the specification.
For Unstructured type PDU Session, the default QoS rule does not contain a Packet Filter Set and in this case the default QoS rule defines the treatment of all packets in the PDU Session.
As long as the default QoS rule does not contain a Packet Filter Set or contains a Packet Filter Set that allows all UL packets, Reflective QoS should not be applied for the QoS Flow which the default QoS rule is associated with and the RQA should not be sent for this QoS Flow. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.5 QoS Flow mapping | The SMF performs the binding of PCC rules to QoS Flows based on the QoS and service requirements (as defined in TS 23.503 [45]). The SMF assigns the QFI for a new QoS Flow and derives its QoS profile, corresponding UPF instructions and QoS Rule(s) from the PCC rule(s) bound to the QoS Flow and other information provided by the PCF.
When applicable, the SMF provides the following information for the QoS Flow to the (R)AN:
- QFI;
- QoS profile as described in clause 5.7.1.2.
- optionally, Alternative QoS Profile(s) as described in clause 5.7.1.2a;
For each PCC rule bound to a QoS Flow, the SMF provides the following information to the UPF enabling classification, bandwidth enforcement and marking of User Plane traffic (the details are described in clause 5.8):
- a DL PDR containing the DL part of the SDF template;
- an UL PDR containing the UL part of the SDF template;
NOTE 1: If a DL PDR for an bidirectional SDF is associated with a QoS Flow other than the one associated with the default QoS rule and the UE has not received any instruction to use this QoS Flow for the SDF in uplink direction (i.e. neither a corresponding QoS rule is sent to the UE nor the Reflective QoS Indication is set in the PCC rule), it means that the UL PDR for the same SDF has to be associated with the QoS Flow associated with the default QoS rule.
- the PDR precedence value (see clause 5.7.1.9) for both PDRs is set to the precedence value of the PCC rule;
- QoS related information (e.g. MBR for an SDF, GFBR and MFBR for a GBR QoS Flow) as described in clause 5.8.2;
- the corresponding packet marking information (e.g. the QFI, the transport level packet marking value (e.g. the DSCP value of the outer IP header);
- optionally, the Reflective QoS Indication is included in the QER associated with the DL PDR (as described in clause 5.7.5.3).
For each PCC rule bound to a QoS Flow, when applicable, the SMF generates an explicitly signalled QoS rule (see clause 5.7.1.4) according to the following principles and provides it to the UE together with an add operation:
- A unique (for the PDU Session) QoS rule identifier is assigned;
- The QFI in the QoS rule is set to the QFI of the QoS Flow to which the PCC rule is bound;
- The Packet Filter Set of the QoS rule is generated from the UL SDF filters and optionally the DL SDF filters of the PCC rule (but only from those SDF filters that have an indication for being signalled to the UE, as defined in TS 23.503 [45]);
- The QoS rule precedence value is set to the precedence value of the PCC rule for which the QoS rule is generated;
- for a dynamically assigned QFI, the QoS Flow level QoS parameters (e.g. 5QI, GFBR, MFBR, Averaging Window, see TS 24.501 [47]) are signalled to UE in addition to the QoS rule(s) associated to the QoS Flow. The QoS Flow level QoS parameters of an existing QoS Flow may be updated based on the MBR and GBR information received in the PCC rule (MBR and GBR per SDF are however not provided to UE over N1 in the case of more than one SDF) or, if the PCF has not indicated differently, when Notification control or handover related signalling indicates that the QoS parameter the NG-RAN is currently fulfilling for the QoS Flow have changed (see clause 5.7.2.4).
Changes in the binding of PCC rules to QoS Flows as well as changes in the PCC rules or other information provided by the PCF can require QoS Flow changes which the SMF has to provide to (R)AN, UPF and/or UE. In the case of changes in the explicitly signalled QoS rules associated to a QoS Flow, the SMF provides the explicitly signalled QoS rules and their operation (i.e. add/modify/delete) to the UE.
NOTE 2: The SMF cannot provide, update or remove pre-configured QoS rules or UE derived QoS rules.
The principle for classification and marking of User Plane traffic and mapping of QoS Flows to AN resources is illustrated in Figure 5.7.1.5-1.
Figure 5.7.1.5-1: The principle for classification and User Plane marking for QoS Flows and mapping to AN Resources
In DL, incoming data packets are classified by the UPF based on the Packet Filter Sets of the DL PDRs in the order of their precedence (without initiating additional N4 signalling). The UPF conveys the classification of the User Plane traffic belonging to a QoS Flow through an N3 (and N9) User Plane marking using a QFI. The AN binds QoS Flows to AN resources (i.e. Data Radio Bearers of in the case of 3GPP RAN). There is no strict 1:1 relation between QoS Flows and AN resources. It is up to the AN to establish the necessary AN resources that QoS Flows can be mapped to and to release them. The AN shall indicate to the SMF when the AN resources onto which a QoS Flow is mapped are released.
If no matching DL PDR is found, the UPF shall discard the DL data packet.
In UL:
- For a PDU Session of Type IP or Ethernet, the UE evaluates UL packets against the UL Packet Filters in the Packet Filter Set in the QoS rules based on the precedence value of QoS rules in increasing order until a matching QoS rule (i.e. whose Packet Filter matches the UL packet) is found.
- If no matching QoS rule is found, the UE shall discard the UL data packet.
- For a PDU Session of Type Unstructured, the default QoS rule does not contain a Packet Filter Set and allows all UL packets.
NOTE 3: Only the default QoS rule exist for a PDU Session of Type Unstructured.
The UE uses the QFI in the corresponding matching QoS rule to bind the UL packet to a QoS Flow. The UE then binds QoS Flows to AN resources. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.6 DL traffic | The following characteristics apply for processing of DL traffic:
- UPF maps User Plane traffic to QoS Flows based on the PDRs.
- UPF performs Session-AMBR enforcement as specified in clause 5.7.1.8 and performs counting of packets for charging.
- UPF transmits the PDUs of the PDU Session in a single tunnel between 5GC and (R)AN, the UPF includes the QFI in the encapsulation header. In addition, UPF may include an indication for Reflective QoS activation in the encapsulation header.
- UPF performs transport level packet marking in DL on a per QoS Flow basis (optionally taking into account PDU Set Information, if applicable). The UPF uses the transport level packet marking value(s) provided by the SMF (as described in clause 5.8.2.7).
- (R)AN maps PDUs from QoS Flows to access-specific resources based on the QFI and the associated 5G QoS profile, also taking into account the N3 tunnel associated with the DL packet.
NOTE: Packet Filters are not used for the mapping of QoS Flows onto access-specific resources in (R)AN.
- If Reflective QoS applies, the UE creates a new derived QoS rule as defined in clause 5.7.5.2. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.7 UL Traffic | Following characteristics apply for processing of UL traffic:
- UE uses the stored QoS rules to determine mapping between UL User Plane traffic and QoS Flows. UE marks the UL PDU with the QFI of the QoS rule containing the matching Packet Filter and transmits the UL PDUs using the corresponding access specific resource for the QoS Flow based on the mapping provided by (R)AN. For NG-RAN, the UL behaviour is specified in clause 10.5.2 of TS 38.300 [27].
- (R)AN transmits the PDUs over N3 tunnel towards UPF. When passing an UL packet from (R)AN to CN, the (R)AN includes the QFI value, in the encapsulation header of the UL PDU and selects the N3 tunnel.
- (R)AN performs transport level packet marking in the UL on a per QoS Flow basis with a transport level packet marking value that is determined based on the 5QI, the Priority Level (if explicitly signalled) and the ARP priority level of the associated QoS Flow.
- UPF verifies whether QFIs in the UL PDUs are aligned with the QoS Rules provided to the UE or implicitly derived by the UE in the case of Reflective QoS).
- UPF and UE perform Session-AMBR enforcement as specified in clause 5.7.1.8 and the UPF performs counting of packets for charging. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.8 AMBR/MFBR enforcement and rate limitation | UL and DL Session-AMBR (see clause 5.7.2.6) shall be enforced by the UPF, if the UPF receives the Session-AMBR values from the SMF as described in clause 5.8.2.7 and clause 5.8.5.4.
For UL Classifier PDU Sessions, UL and DL Session-AMBR (see clause 5.7.2.6) shall be enforced in the SMF selected UPF that supports the UL Classifier functionality. In addition, the DL Session-AMBR shall be enforced separately in every UPF that terminates the N6 interface (i.e. without requiring interaction between the UPFs) (see clause 5.6.4).
For multi-homed PDU Sessions, UL and DL Session-AMBR shall be enforced in the UPF that supports the Branching Point functionality. In addition, the DL Session-AMBR shall be enforced separately in every UPF that terminates the N6 interface (i.e. without requiring interaction between the UPFs) (see clause 5.6.4).
NOTE: The DL Session-AMBR is enforced in every UPF terminating the N6 interface to reduce unnecessary transport of traffic which may be discarded by the UPF performing the UL Classifier/Branching Point functionality due to the amount of the DL traffic for the PDU Session exceeding the DL Session-AMBR. Discarding DL packets in the UL Classifier/Branching Point could cause erroneous PDU counting for support of charging
For HR-SBO PDU Sessions, if the DL Session AMBR for Offloading is provided as described (see clause 6.7 of TS 23.548 [130]), the DL Session AMBR for Offloading is enforced by UPF that supports the UL Classifier/Branching Point functionality in the VPLMN and additionally separately in every PSA UPF that terminates the N6 interface toward the local part of DN in the VPLMN.
The (R)AN shall enforce UE-AMBR (see clause 5.7.2.6) in UL and DL per UE for Non-GBR QoS Flows.
The UE shall perform UL rate limitation on PDU Session basis for Non-GBR traffic using Session-AMBR, if the UE receives a Session-AMBR.
MBR per SDF is mandatory for GBR QoS Flows but optional for Non-GBR QoS Flows. The MBR is enforced in the UPF.
The MFBR is enforced in the UPF in the Downlink for GBR QoS Flows. The MFBR is enforced in the (R)AN in the Downlink and Uplink for GBR QoS Flows. For non-3GPP access, the UE should enforce MFBR in the Uplink for GBR QoS Flows.
The QoS control for Unstructured PDUs is performed at the PDU Session level and in this Release of the specification there is only support for maximum of one 5G QoS Flow per PDU Session of Type Unstructured.
When a PDU Session is set up for transferring unstructured PDUs, SMF provides the QFI which will be applied to any packet of the PDU Session to the UPF and UE. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.9 Precedence Value | The QoS rule precedence value and the PDR precedence value determine the order in which a QoS rule or a PDR, respectively, shall be evaluated. The evaluation of the QoS rules or PDRs is performed in increasing order of their precedence value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.10 UE-Slice-MBR enforcement and rate limitation | If a supporting NG-RAN receives for a UE a UE-Slice-MBR (see clause 5.7.2.6) for an S-NSSAI from the AMF, the NG-RAN shall apply this UE-Slice-MBR for all PDU Sessions of that UE corresponding to the S-NSSAI which have an active user plane if feasible. In particular, the NG-RAN shall enforce this UE-Slice-MBR as follows:
1) Whenever a request for a GBR QoS Flow establishment or modification is received, the NG-RAN admission control shall ensure that the sum of the GFBR values of the admitted GBR QoS Flows is not exceeding the UE-Slice-MBR and, if the QoS Flow cannot be admitted, the NG-RAN shall reject the establishment/modification of the QoS Flow.
NOTE: If the UE-Slice-MBR would be exceeded by a new/modified GBR QoS Flow, the NG-RAN determines whether the new/modified GBR QoS Flow can pre-empt any existing GBR QoS Flow of the UE's PDU Session(s) corresponding to the same S-NSSAI based on their ARP values (as per clause 5.7.2.2), e.g. to support certain priority services (e.g. MPS). If this is not possible, the NG-RAN can reject the establishment/modification of the QoS Flow.
2) The NG-RAN shall ensure that the aggregated bitrate across all GBR and Non-GBR QoS Flows belonging to those PDU Sessions is not exceeding the UE-Slice-MBR, while always guaranteeing the GFBR of every GBR QoS Flow of those PDU Sessions as described in clause 5.7.2.5. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.1.11 QoS aspects of home-routed roaming | In the case of home-routed roaming, the V-SMF may apply VPLMN policies related with the SLA negotiated with the HPLMN or with QoS values supported by the VPLMN. Such policies may result in a situation that the V-SMF does not accept the PDU Session or does not accept some of the QoS Flows requested by the H-SMF.
QoS constraints represent the QoS that the VPLMN can accept for the QoS Flow associated with the default QoS rule and the PDU Session based on SLA or based on QoS values supported by the VPLMN. The QoS constraints may contain 5QI(s), 5QI Priority Level and ARP for the QoS Flow associated with the default QoS rule and highest Session-AMBR accepted by the VPLMN.
NOTE 1: For this Release of the specification, QoS constraints apply only to the non-GBR default QoS Flow.
NOTE 2: Additionally, at inter-PLMN scenarios, the VPLMN policies can result in re-mapping of QoS values of non-default QoS Flows e.g. in case the V-SMF receives an operator-specific 5QI value, whether/how the remapping is done is implementation specific according to the local policy and according to an inter-PLMN agreement between the involved operators.
NOTE 3: If remapping at inter-PLMN mobility between HPLMN and VPLMN is to be performed, the V-SMF chooses a 5QI value based on inter-PLMN agreement between the VPLMN and the HPLMN and updates the QoS Flows accordingly. The remapping can result in service degradation. At inter-PLMN mobility between two VPLMNs, the source V-SMF provides the 5QI values used in the source VPLMN to the target VPLMN. If the source VPLMN provides a 5QI value which is not part of the inter-PLMN agreement between the target VPLMN and the HPLMN, the target VPLMN can choose a 5QI value based on local policy and updates the QoS Flows accordingly using the procedure in clause 4.3.3.3 of TS 23.502 [3].
NOTE 4: If remapping at session management procedure in roaming case is to be performed, the V-SMF chooses a 5QI value based on inter-PLMN agreement between the VPLMN and the HPLMN and updates the QoS Flows accordingly. The remapping can result in service degradation.
NOTE 5: The VPLMN can provide multiple 5QI values, as constraints for the default QoS rule to HPLMN that can be accepted by VPLMN, based on inter-PLMN agreement. For other parameters (e.g. 5QI Priority level, ARP), the VPLMN provides the highest limit as constraint.
At PDU Session Establishment for home-routed roaming, to reduce the risk of PDU Session establishment failure due to QoS from the HPLMN not being compliant with SLA, the V-SMF may provide the VPLMN local policy in QoS constraints to the H-SMF as specified in clause 4.3.2.2.2 of TS 23.502 [3].
For intra-5GS mobility with V-SMF insertion or V-SMF change (e.g. inter-PLMN mobility), as specified in clause 4.23 of TS 23.502 [3], the new/target V-SMF may validate the currently applied QoS against the QoS constraints. The new/target V-SMF provides QoS constraints to the H-SMF during the mobility procedure. The new/target V-SMF may temporarily accept a higher QoS even if the currently applied QoS exceeds the QoS constraints. Alternatively, for the QoS parameters related with the QoS constraints, the V-SMF may locally downgrade these values before providing the corresponding QoS profiles to 5G AN. The V-SMF may decide to release the PDU Session if the HPLMN does not provide updated QoS compliant with the QoS constraints after the mobility procedure.
For IMS voice service (e.g. the IMS DNN defined by the GSMA), the V-SMF, based on local policy, may override the ARP received from HPLMN over N16 if the ARP indicates priority not in line with the local policy in VPLMN. The ARP override in the serving PLMN applies to both the QoS Flow associated with the default QoS rule and the QoS Flows for IMS voice, to apply the same allocation and retention priority for all users (i.e. roamers and non-roamers). For MPS (clause 5.16.5), the same allocation and retention priority is applied to all MPS service users (i.e. roamers and non-roamers), when roaming agreements are in place and where regulatory requirements apply. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2 5G QoS Parameters | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.1 5QI | A 5QI is a scalar that is used as a reference to 5G QoS characteristics defined in clause 5.7.4, i.e. access node-specific parameters that control QoS forwarding treatment for the QoS Flow (e.g. scheduling weights, admission thresholds, queue management thresholds, link layer protocol configuration, etc.).
Standardized 5QI values have one-to-one mapping to a standardized combination of 5G QoS characteristics as specified in Table 5.7.4-1.
The 5G QoS characteristics for pre-configured 5QI values are pre-configured in the AN.
Standardized or pre-configured 5G QoS characteristics, are indicated through the 5QI value and are not signalled on any interface, unless certain 5G QoS characteristics are modified as specified in clauses 5.7.3.3, 5.7.3.4, 5.7.3.6 and 5.7.3.7.
The 5G QoS characteristics for QoS Flows with dynamically assigned 5QI are signalled as part of the QoS profile.
NOTE: On N3, each PDU (i.e. in the tunnel used for the PDU Session) is associated with one 5QI via the QFI carried in the encapsulation header. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.2 ARP | The QoS parameter ARP contains information about the priority level, the pre-emption capability and the pre-emption vulnerability. This allows deciding whether a QoS Flow establishment/modification/handover may be accepted or needs to be rejected in the case of resource limitations (typically used for admission control of GBR traffic). It may also be used to decide which existing QoS Flow to pre-empt during resource limitations, i.e. which QoS Flow to release to free up resources.
The ARP priority level defines the relative importance of a QoS Flow. The range of the ARP priority level is 1 to 15 with 1 as the highest priority.
The ARP priority levels 1-8 should only be assigned to QoS Flows for services that are authorized to receive prioritized treatment within an operator domain (i.e. that are authorized by the serving network). The ARP priority levels 9-15 may be assigned to QoS Flows for services that are authorized by the home network and thus applicable when a UE is roaming.
NOTE: This ensures that future releases may use ARP priority level 1-8 to indicate e.g. emergency and other priority services within an operator domain in a backward compatible manner. This does not prevent the use of ARP priority level 1-8 in roaming situation in the case that appropriate roaming agreements exist that ensure a compatible use of these priority levels.
The ARP pre-emption capability defines whether a QoS Flow may get resources that were already assigned to another QoS Flow with a lower priority. The ARP pre-emption vulnerability defines whether a QoS Flow may lose the resources assigned to it in order to admit a QoS Flow with higher priority. The ARP pre-emption capability and the ARP pre-emption vulnerability shall be either set to 'enabled' or 'disabled'.
The ARP pre-emption vulnerability of the QoS Flow which the default QoS rule is associated with should be set appropriately to minimize the risk of a release of this QoS Flow.
The details of how the SMF sets the ARP for a QoS Flow are further described in clause 5.7.2.7. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.3 RQA | The Reflective QoS Attribute (RQA) is an optional parameter which indicates that certain traffic (not necessarily all) carried on this QoS Flow is subject to Reflective QoS. Only when the RQA is signalled for a QoS Flow, the (R)AN enables the transfer of the RQI for AN resource corresponding to this QoS Flow. The RQA may be signalled to NG-RAN via the N2 reference point at UE context establishment in NG-RAN and at QoS Flow establishment or modification. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.4 Notification control | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.4.1 General | The QoS Parameter Notification control indicates to the NG-RAN that notifications of "GFBR can no longer (or can again) be guaranteed" are requested when the NG-RAN determines that the GFBR, the PDB or the PER of the QoS profile cannot be fulfilled (or can be fulfilled again) for a QoS Flow (during the lifetime of the QoS Flow) and that the QoS Flow should be kept while the NG-RAN is not fulfilling the requested QoS profile. Notification control may be used for a GBR QoS Flow if the application traffic is able to adapt to the change in the QoS (e.g. if the AF is capable to trigger rate adaptation).
The NG-RAN determination whether the GFBR, the PDB and the PER can be fulfilled or not, is done under consideration of the QoS parameters Averaging Window and MDBV (for a GBR QoS Flow using the Delay-critical resource type), which are either provided explicitly as part of the QoS profile or implicitly via the standardized 5QI QoS characteristics defined in clause 5.7.4.
The SMF shall only enable Notification control when the QoS Notification Control parameter is set in the PCC rule (received from the PCF) that is bound to the QoS Flow. The Notification control parameter is signalled to the NG-RAN as part of the QoS profile.
If the NG-RAN has received the PDU Set QoS parameters PSDB and PSER for UL and/or DL (as described in clause 5.7.7) and PDU Set QoS handling is applied in the respective direction(s), the NG-RAN uses the PSDB and PSER instead of the PDB and PER in the respective direction(s) for the actions described in all (sub) clauses of clause 5.7.2.4. If the PDU Set QoS handling, PSER and PSDB are applied in one direction only, the NG-RAN still uses the PDB and PER in the other direction for the actions described in all (sub) clauses of clause 5.7.2.4.
5.7.2.4.1a Notification Control without Alternative QoS Profiles
If, for a given GBR QoS Flow, Notification control is enabled and the NG-RAN determines that the GFBR, the PDB or the PER of the QoS profile cannot be fulfilled, NG-RAN shall send a notification towards SMF that the "GFBR can no longer be guaranteed". Furthermore, the NG-RAN shall keep the QoS Flow (i.e. while the NG-RAN is not fulfilling the requested QoS profile for this QoS Flow), unless specific conditions at the NG-RAN require the release of the NG-RAN resources for this GBR QoS Flow, e.g. due to Radio link failure or RAN internal congestion. The NG-RAN should try to fulfil the GFBR, the PDB and the PER of the QoS profile again.
NOTE 1: NG-RAN can decide that the "GFBR can no longer be guaranteed" based on, e.g. measurements like queuing delay or system load.
Upon receiving a notification from the NG-RAN that the "GFBR can no longer be guaranteed", the SMF may forward the notification to the PCF, see TS 23.503 [45].
When the NG-RAN determines that the GFBR, the PDB and the PER of the QoS profile can be fulfilled again for a QoS Flow (for which a notification that the "GFBR can no longer be guaranteed" has been sent), the NG-RAN shall send a notification, informing the SMF that the "GFBR can again be guaranteed" and the SMF may forward the notification to the PCF, see TS 23.503 [45]. The NG-RAN shall send a subsequent notification that the "GFBR can no longer be guaranteed" whenever necessary.
NOTE 2: It is assumed that NG-RAN implementation will apply some hysteresis before determining that the "GFBR can again be guaranteed" and therefore a frequent signalling of "GFBR can again be guaranteed" followed by "GFBR can no longer be guaranteed" is not expected.
NOTE 3: If the QoS Flow is modified, the NG-RAN restarts the check whether the "GFBR can no longer be guaranteed" according to the updated QoS profile. If the Notification control parameter is not included in the updated QoS profile, the Notification control is disabled.
During a handover, the Source NG-RAN does not inform the Target NG-RAN about whether the Source NG-RAN has sent a notification for a QoS Flow that the "GFBR can no longer be guaranteed". The Target NG-RAN performs admission control rejecting any QoS Flows for which resources cannot be permanently allocated. The accepted QoS Flows are included in the N2 Path Switch Request or N2 Handover Request Acknowledge message from the NG-RAN to the AMF. The SMF shall interpret the fact that a QoS Flow is listed as transferred QoS Flow in the Nsmf_PDUSession_UpdateSMContext Request received from the AMF as a notification that "GFBR can again be guaranteed" for this QoS Flow unless the SMF is also receiving a reference to an Alternative QoS Profile for this QoS Flow (which is described in clause 5.7.2.4.2). After the handover is successfully completed, the Target NG-RAN shall send a subsequent notification that the "GFBR can no longer be guaranteed" for such a QoS Flow whenever necessary. If the SMF has previously notified the PCF that the "GFBR can no longer be guaranteed" and the SMF does not receive an explicit notification that the "GFBR can no longer be guaranteed" for that QoS Flow from the Target NG-RAN within a configured time, the SMF shall notify the PCF that the "GFBR can again be guaranteed".
If the PDU Set QoS handling is applied in UL and/or DL, the NG-RAN uses the PDU Set QoS parameters PSDB and PSER for UL and/or DL instead of PDB and PER in the respective direction(s) to determine whether "GFBR can no longer be guaranteed" or "GFBR can again be guaranteed".
If the NG-RAN determines that only in one direction the GFBR, the PDB or the PER cannot be fulfilled, the NG-RAN shall indicate that "GFBR can no longer be guaranteed" together with the corresponding direction.
5.7.2.4.1b Notification control with Alternative QoS Profiles
If, for a given GBR QoS Flow, Notification control is enabled and the NG-RAN has received a list of Alternative QoS Profile(s) for this QoS Flow and supports the Alternative QoS Profile handling, the following shall apply:
1) If the NG-RAN determines that the GFBR, the PDB or the PER of the QoS profile cannot be fulfilled, NG-RAN shall send a notification towards SMF that the "GFBR can no longer be guaranteed". Before sending a notification that the "GFBR can no longer be guaranteed" towards the SMF, the NG-RAN shall check whether the GFBR, the PDB and the PER that the NG-RAN currently fulfils match any of the Alternative QoS Profile(s) in the indicated priority order. If there is a match, the NG-RAN shall indicate the reference to the matching Alternative QoS Profile with the highest priority together with the notification to the SMF.
If there is no match, the NG-RAN shall send a notification that the "GFBR can no longer be guaranteed" towards the SMF indicating that the lowest Alternative QoS Profile cannot be fulfilled (unless specific conditions at the NG-RAN require the release of the NG-RAN resources for this GBR QoS Flow, e.g. due to Radio link failure or RAN internal congestion).
2) If a notification that the "GFBR can no longer be guaranteed" has been sent to the SMF and the NG-RAN determines that the currently fulfilled GFBR, PDB or PER are different (better or worse) from the situation indicated in the last notification, the NG-RAN shall send a notification (i.e. "GFBR can no longer be guaranteed" or "GFBR can again be guaranteed") to the SMF and indicate the current situation (unless specific conditions at the NG-RAN require the release of the NG-RAN resources for this GBR QoS Flow, e.g. due to Radio link failure or RAN internal congestion).
NOTE 1: The current situation is either that the QoS Profile can be fulfilled (which is implicitly indicated by the "GFBR can again be guaranteed" notification itself), that a different Alternative QoS Profile can be fulfilled, or that the lowest priority Alternative QoS Profile cannot be fulfilled.
3)- The NG-RAN should always try to fulfil the QoS profile and, if this is not possible, any Alternative QoS Profile that has higher priority.
NOTE 2: In order to avoid a too frequent signalling to the SMF, it is assumed that NG-RAN implementation can apply hysteresis (e.g. via a configurable time interval) before notifying the SMF that the currently fulfilled values match the QoS Profile or a different Alternative QoS Profile of higher priority. It is also assumed that the PCF has ensured that the QoS values within the different Alternative QoS Profile(s) are not too close to each other.
4) Upon receiving a notification from the NG-RAN, the SMF may inform the PCF. If it does so, the SMF shall indicate the currently fulfilled situation to the PCF. See TS 23.503 [45].
5) If the PCF has not indicated differently, the SMF uses NAS signalling (that is sent transparently through the RAN) to inform the UE about changes in the QoS parameters (i.e. 5QI, GFBR, MFBR) that the NG-RAN is currently fulfilling for the QoS Flow after Notification control has occurred. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.4.2 Usage of Notification control with Alternative QoS Profiles at handover | During handover, the prioritized list of Alternative QoS Profile(s) (if available) is provided to the Target NG-RAN per QoS Flow in addition to the QoS profile. If the Target NG-RAN is not able to guarantee the GFBR, the PDB and the PER included in the QoS profile and if Alternative QoS Profiles are provided to the Target NG-RAN and the Target NG-RAN supports Alternative QoS Profiles, the Target NG-RAN checks whether the GFBR, the PDB and the PER values that it can fulfil match any of the Alternative QoS Profile(s) taking the priority order into account. If there is a match between one of the Alternative QoS Profiles and the GFBR, the PDB and the PER values that Target NG-RAN can fulfil, the Target NG-RAN shall accept the QoS Flow and indicate the reference to that Alternative QoS Profile to the Source NG-RAN.
If there is no match to any Alternative QoS Profile, the Target NG-RAN rejects QoS Flows for which the Target NG-RAN is not able to guarantee the GFBR, the PDB and the PER included in the QoS profile.
After the handover is completed and a QoS Flow has been accepted by the Target NG-RAN based on an Alternative QoS Profile, the Target NG-RAN shall treat this QoS Flow in the same way as if it had sent a notification that the "GFBR can no longer be guaranteed" with a reference to that Alternative QoS Profile to the SMF (as described in clause 5.7.2.4.1b).
If a QoS Flow has been accepted by the Target NG-RAN based on an Alternative QoS Profile, the reference to the matching Alternative QoS Profile is provided from the Target NG-RAN to the AMF (which forwards the message to the SMF) during the Xn and N2 based handover procedures as described in TS 23.502 [3]. After the handover is completed successfully, the SMF shall send a notification to the PCF that the "GFBR can no longer be guaranteed" for a QoS Flow (see TS 23.503 [45] for details) if the SMF has received a reference to an Alternative QoS Profile and this reference indicates a change in the previously notified state of this QoS Flow. If the PCF has not indicated differently, the SMF shall also use NAS signalling (that is sent transparently through the RAN) to inform the UE about the QoS parameters (i.e. 5QI, GFBR, MFBR) corresponding to the new state of the QoS Flow.
NOTE: A state change for the QoS Flow comprises a change from QoS profile fulfilled to Alternative QoS Profile fulfilled as well as the state change between fulfilled Alternative QoS Profiles.
If a QoS Flow has been accepted by the Target NG-RAN based on the QoS Profile, the SMF shall interpret the fact that a QoS Flow is listed as transferred QoS Flow in the message received from the AMF as a notification that "GFBR can again be guaranteed" for this QoS Flow. After the handover is successfully completed, the Target NG-RAN performs as described in clause 5.7.2.4.1b. If the SMF has previously notified the PCF that the "GFBR can no longer be guaranteed" and the SMF does not receive an explicit notification that the "GFBR can no longer be guaranteed" for that QoS Flow from the Target NG-RAN within a configured time, the SMF shall notify the PCF that the "GFBR can again be guaranteed".
If a QoS Flow has been accepted by the Target NG-RAN and SMF did not receive from the Target NG-RAN a reference to any Alternative QoS Profile and the SMF has previously informed the UE about QoS parameters corresponding to any of the Alternative QoS Profile(s), the SMF shall use NAS signalling to inform the UE about the QoS parameters corresponding to the QoS Profile. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.4.3 Usage of Notification control with Alternative QoS Profiles during QoS Flow establishment and modification | During QoS Flow establishment and modification, a prioritized list of Alternative QoS Profile(s) can be provided to the NG-RAN for the QoS Flow in addition to the QoS profile. If the NG-RAN is not able to guarantee the GFBR, the PDB and the PER included in the QoS profile and if Alternative QoS Profiles are provided to the NG-RAN and the NG-RAN supports Alternative QoS Profiles, the NG-RAN shall check whether the GFBR, the PDB and the PER values that it can fulfil match at least one of the Alternative QoS Profile(s) taking the priority order into account. If there is a match between one of the Alternative QoS Profiles and the GFBR, the PDB and the PER values that the NG-RAN can fulfil, the NG-RAN shall accept the QoS Flow and indicate the reference to that Alternative QoS Profile to the SMF. If there is no match to any Alternative QoS Profile, the NG-RAN shall reject the QoS Flow establishment or modification.
After a successful QoS Flow establishment or modification during which the NG-RAN indicated that the currently fulfilled QoS matches one of the Alternative QoS Profiles, the NG-RAN shall treat this QoS Flow in the same way as if it had sent a notification that the "GFBR can no longer be guaranteed" with a reference to that Alternative QoS Profile to the SMF (as described in clause 5.7.2.4.1b).
If the SMF has received a reference to an Alternative QoS Profile during QoS Flow establishment and modification the SMF may inform the PCF about it (as described in TS 23.503 [45]).
If the PCF has not indicated differently, the SMF shall use NAS signalling (that is sent transparently through the RAN) to inform the UE about the QoS parameters (i.e. 5QI, GFBR, MFBR) corresponding to the referenced Alternative QoS Profile. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.5 Flow Bit Rates | For GBR QoS Flows only, the following additional QoS parameters exist:
- Guaranteed Flow Bit Rate (GFBR) - UL and DL;
- Maximum Flow Bit Rate (MFBR) -- UL and DL.
The GFBR denotes the bit rate that is guaranteed to be provided by the network to the QoS Flow over the Averaging Time Window. The MFBR limits the bit rate to the highest bit rate that is expected by the QoS Flow (e.g. excess traffic may get discarded or delayed by a rate shaping or policing function at the UE, RAN, UPF). Bit rates above the GFBR value and up to the MFBR value, may be provided with relative priority determined by the Priority Level of the QoS Flows (see clause 5.7.3.3).
GFBR and MFBR are signalled to the (R)AN in the QoS Profile and signalled to the UE as QoS Flow level QoS parameter (as specified in TS 24.501 [47]) for each individual QoS Flow.
NOTE 1: The GFBR is recommended as the lowest acceptable service bitrate where the service will survive.
NOTE 2: For each QoS Flow of Delay-critical GBR resource type, the SMF can ensure that the GFBR of the QoS Flow can be achieved with the MDBV of the QoS Flow using the QoS Flow binding functionality described in clause 6.1.3.2.4 of TS 23.503 [45].
NOTE 3: The network can set MFBR larger than GFBR for a particular QoS Flow based on operator policy and the knowledge of the end point capability, i.e. support of rate adaptation at application / service level. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.6 Aggregate Bit Rates | Each PDU Session of a UE is associated with the following aggregate rate limit QoS parameter:
- per Session Aggregate Maximum Bit Rate (Session-AMBR).
The Session-AMBR is signalled to the appropriate UPF entity/ies to the UE and to the (R)AN (to enable the calculation of the UE-AMBR). The Session-AMBR limits the aggregate bit rate that can be expected to be provided across all Non-GBR QoS Flows for a specific PDU Session. The Session-AMBR is measured over an AMBR averaging window which is a standardized value. The Session-AMBR is not applicable to GBR QoS Flows.
Each UE is associated with the following aggregate rate limit QoS parameter:
- per UE Aggregate Maximum Bit Rate (UE-AMBR).
The UE-AMBR limits the aggregate bit rate that can be expected to be provided across all Non-GBR QoS Flows of a UE. Each (R)AN shall set its UE-AMBR to the sum of the Session-AMBR of all PDU Sessions with active user plane to this (R)AN up to the value of the UE-AMBR received from AMF. The UE-AMBR is a parameter provided to the (R)AN by the AMF based on the value of the subscribed UE-AMBR retrieved from UDM or the dynamic serving network UE-AMBR retrieved from PCF (e.g. for roaming subscriber). The AMF provides the UE-AMBR provided by PCF to (R)AN if available. The UE-AMBR is measured over an AMBR averaging window which is a standardized value. The UE-AMBR is not applicable to GBR QoS Flows.
Each group of PDU Sessions of the UE for the same slice (S-NSSAI) may be associated with the following aggregate rate limit QoS parameter:
- per UE per Slice-Maximum Bit Rate (UE-Slice-MBR).
The UE-Slice-MBR limits the aggregate bit rate that can be expected to be provided across all GBR and Non-GBR QoS Flows corresponding to PDU Sessions of the UE for the same slice (S-NSSAI) which have an active user plane. Each supporting NG-RAN shall set its UE-Slice-MBR to the sum of the Session-AMBR and MFBR for GBR QoS Flows of all PDU Sessions corresponding to the slice (S-NSSAI) with active user plane to this NG-RAN up to the value of the UE-Slice-MBR corresponding to the slice (S-NSSAI) received from AMF. The UE-Slice-MBR is measured over an AMBR averaging window which is a standardized value. The UE-Slice-MBR is an optional parameter provided to the NG-RAN by the AMF as described in clause 5.15.13.
NOTE: The AMBR averaging window is only applied to Session-AMBR, UE-AMBR and UE-Slice-MBR measurement and the AMBR averaging windows for Session-AMBR and UE-AMBR are standardised to the same value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.7 Default values | For each PDU Session Setup, the SMF retrieves the subscribed Session-AMBR values as well as the subscribed default values for the 5QI and the ARP and optionally, the 5QI Priority Level, from the UDM. The subscribed default 5QI value shall be a Non-GBR 5QI from the standardized value range.
NOTE 1: The 5QI Priority Level can be added to the subscription information to achieve an overwriting of the standardized or pre-configured 5QI Priority Level e.g. in scenarios where dynamic PCC is not deployed or the PCF is unavailable or unreachable.
The SMF may change the subscribed values for the default 5QI and the ARP and if received, the 5QI Priority Level, based on interaction with the PCF as described in TS 23.503 [45] or, if dynamic PCC is not deployed, based on local configuration, to set QoS parameters for the QoS Flow associated with the default QoS rule.
For QoS Flow(s) of the PDU Session other than the QoS Flow associated with the default QoS rule, the SMF shall set the ARP priority level, the ARP pre-emption capability and the ARP pre-emption vulnerability to the respective values in the PCC rule(s) bound to that QoS Flow (as described in TS 23.503 [45]). If dynamic PCC is not deployed, the SMF shall set the ARP priority level, the ARP pre-emption capability and the ARP pre-emption vulnerability based on local configuration.
NOTE 2: The local configuration in the SMF can e.g. make use of the subscribed value for the ARP priority level and apply locally configured values for the ARP pre-emption capability and ARP pre-emption vulnerability.
If dynamic PCC is not deployed, the SMF can have a DNN based configuration to enable the establishment of a GBR QoS Flow as the QoS Flow that is associated with the default QoS rule. This configuration contains a standardized GBR 5QI as well as GFBR and MFBR for UL and DL.
NOTE 3: Interworking with EPS is not possible for a PDU Session with a GBR QoS Flow as the QoS Flow that is associated with the default QoS rule.
The SMF may change the subscribed Session-AMBR values (for UL and/or DL), based on interaction with the PCF as described in TS 23.503 [45] or, if dynamic PCC is not deployed, based on local configuration, to set the Session-AMBR values for the PDU Session. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.8 Maximum Packet Loss Rate | The Maximum Packet Loss Rate (UL, DL) indicates the maximum rate for lost packets of the QoS Flow that can be tolerated in the uplink and downlink direction. This is provided to the QoS Flow if it is compliant to the GFBR
NOTE: In this Release of the specification, the Maximum Packet Loss Rate (UL, DL) can only be provided for a GBR QoS Flow belonging to voice media. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.2.9 Wireline access network specific 5G QoS parameters | QoS parameters that are applicable only for or wireline access networks (W-5GAN) are specified in TS 23.316 [84]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3 5G QoS characteristics | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.1 General | This clause specifies the 5G QoS characteristics associated with 5QI. The characteristics describe the packet forwarding treatment that a QoS Flow receives edge-to-edge between the UE and the UPF in terms of the following performance characteristics:
1 Resource type (Non-GBR, GBR, Delay-critical GBR);
2 Priority Level;
3 Packet Delay Budget (including Core Network Packet Delay Budget);
4 Packet Error Rate;
5 Averaging window (for GBR and Delay-critical GBR resource type only);
6 Maximum Data Burst Volume (for Delay-critical GBR resource type only).
The 5G QoS characteristics should be understood as guidelines for setting node specific parameters for each QoS Flow e.g. for 3GPP radio access link layer protocol configurations.
Standardized or pre-configured 5G QoS characteristics, are indicated through the 5QI value and are not signalled on any interface, unless certain 5G QoS characteristics are modified as specified in clauses 5.7.3.3, 5.7.3.4, 5.7.3.6 and 5.7.3.7.
NOTE: As there are no default values specified, pre-configured 5G QoS characteristics have to include all of the characteristics listed above.
Signalled 5G QoS characteristics are provided as part of the QoS profile and shall include all of the characteristics listed above. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.2 Resource Type | The resource type determines if dedicated network resources related to a QoS Flow-level Guaranteed Flow Bit Rate (GFBR) value are permanently allocated (e.g. by an admission control function in a radio base station).
GBR QoS Flows are therefore typically authorized "on demand" which requires dynamic policy and charging control. A GBR QoS Flow uses either the GBR resource type or the Delay-critical GBR resource type. The definition of PDB and PER are different for GBR and Delay-critical GBR resource types and the MDBV parameter applies only to the Delay-critical GBR resource type.
A Non-GBR QoS Flow may be pre-authorized through static policy and charging control. A Non-GBR QoS Flow uses only the Non-GBR resource type. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.3 Priority Level | The Priority Level associated with 5G QoS characteristics indicates a priority in scheduling resources among QoS Flows. The lowest Priority Level value corresponds to the highest priority.
The Priority Level shall be used to differentiate between QoS Flows of the same UE and it shall also be used to differentiate between QoS Flows from different UEs.
In the case of congestion, when all QoS requirements cannot be fulfilled for one or more QoS Flows, the Priority Level shall be used to select for which QoS Flows the QoS requirements are prioritised such that a QoS Flow with Priority Level value N is priorized over QoS Flows with higher Priority Level values (i.e. N+1, N+2, etc).In the case of no congestion, the Priority Level should be used to define the resource distribution between QoS Flows. In addition, the scheduler may prioritize QoS Flows based on other parameters (e.g. resource type, radio condition) in order to optimize application performance and network capacity.
Every standardized 5QI is associated with a default value for the Priority Level -specified in QoS characteristics Table 5.7.4.1). Priority Level may also be signalled together with a standardized 5QI to the -R)AN and if it is received, it shall be used instead of the default value.
Priority Level may also be signalled together with a pre-configured 5QI to the (R)AN and if it is received, it shall be used instead of the pre-configured value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.4 Packet Delay Budget | The Packet Delay Budget (PDB) defines an upper bound for the time that a packet may be delayed between the UE and the N6 termination point at the UPF. The PDB applies to the DL packet received by the UPF over the N6 interface and to the UL packet sent by the UE. For a certain 5QI the value of the PDB is the same in UL and DL. In the case of 3GPP access, the PDB is used to support the configuration of scheduling and link layer functions (e.g. the setting of scheduling priority weights and HARQ target operating points). For GBR QoS Flows using the Delay-critical resource type, a packet delayed more than PDB is counted as lost if the data burst is not exceeding the MDBV within the period of PDB and the QoS Flow is not exceeding the GFBR. For GBR QoS Flows with GBR resource type not exceeding GFBR, 98 percent of the packets shall not experience a delay exceeding the 5QI's PDB.
The 5G Access Network Packet Delay Budget (5G-AN PDB) is determined by subtracting a static value for the Core Network Packet Delay Budget (CN PDB), which represents the delay between any N6 termination point at the UPF (for any UPF that may possibly be selected for the PDU Session) and the 5G-AN from a given PDB.
NOTE 1: For a standardized 5QI, the static value for the CN PDB is specified in the QoS characteristics Table 5.7.4-1.
NOTE 2: For a non-standardized 5QI, the static value for the CN PDB is homogeneously configured in the network.
For GBR QoS Flows using the Delay-critical resource type, in order to obtain a more accurate delay budget PDB available for the NG-RAN, a dynamic value for the CN PDB, which represents the delay between the UPF terminating N6 for the QoS Flow and the 5G-AN, can be used. If used for a QoS Flow, the NG-RAN shall apply the dynamic value for the CN PDB instead of the static value for the CN PDB (which is only related to the 5QI). Different dynamic value for CN PDB may be configured per uplink and downlink direction.
NOTE 3: The configuration of transport network on CN tunnel can be different per UL and DL, which can be different value for CN PDB per UL and DL.
NOTE 4: It is expected that the UPF deployment ensures that the dynamic value for the CN PDB is not larger than the static value for the CN PDB. This avoids that the functionality that is based on the 5G-AN PDB (e.g. MDBV, NG-RAN scheduler) has to handle an unexpected value.
The dynamic value for the CN PDB of a Delay-critical GBR 5QI may be configured in the network in two ways:
- Configured in each NG-RAN node, based on a variety of inputs such as different IP address(es) or TEID range of UPF terminating the N3 tunnel and based on different combinations of PSA UPF to NG-RAN under consideration of any potential I-UPF, etc;
- Configured in the SMF, based on different combinations of PSA UPF to NG-RAN under consideration of any potential I-UPF. The dynamic value for the CN PDB for a particular QoS Flow shall be signalled to NG-RAN (during PDU Session Establishment, PDU Session Modification, Xn/N2 handover and the Service Request procedures) when the QoS Flow is established or the dynamic value for the CN PDB of a QoS Flow changes, e.g. when an I-UPF is inserted by the SMF.
If the NG-RAN node is configured locally with a dynamic value for the CN PDB for a Delay-critical GBR 5QI and receives a different value via N2 signalling for a QoS Flow with the same 5QI, local configuration in RAN node determines which value takes precedence.
Services using a GBR QoS Flow and sending at a rate smaller than or equal to the GFBR can in general assume that congestion related packet drops will not occur.
NOTE 5: Exceptions (e.g. transient link outages) can always occur in a radio access system which may then lead to congestion related packet drops. Packets surviving congestion related packet dropping may still be subject to non-congestion related packet losses (see PER below).
Services using Non-GBR QoS Flows should be prepared to experience congestion-related packet drops and delays. In uncongested scenarios, 98 percent of the packets should not experience a delay exceeding the 5QI's PDB.
The PDB for Non-GBR and GBR resource types denotes a "soft upper bound" in the sense that an "expired" packet, e.g. a link layer SDU that has exceeded the PDB, does not need to be discarded and is not added to the PER. However, for a Delay-critical GBR resource type, packets delayed more than the PDB are added to the PER and can be discarded or delivered depending on local decision. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.5 Packet Error Rate | The Packet Error Rate (PER) defines an upper bound for the rate of PDUs (e.g. IP packets) that have been processed by the sender of a link layer protocol (e.g. RLC in RAN of a 3GPP access) but that are not successfully delivered by the corresponding receiver to the upper layer (e.g. PDCP in RAN of a 3GPP access). Thus, the PER defines an upper bound for a rate of non-congestion related packet losses. The purpose of the PER is to allow for appropriate link layer protocol configurations (e.g. RLC and HARQ in RAN of a 3GPP access). For every 5QI the value of the PER is the same in UL and DL. For GBR QoS Flows with Delay-critical GBR resource type, a packet which is delayed more than PDB is counted as lost and included in the PER unless the data burst is exceeding the MDBV within the period of PDB or the QoS Flow is exceeding the GFBR. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.6 Averaging Window | Each GBR QoS Flow shall be associated with an Averaging window. The Averaging window represents the duration over which the GFBR and MFBR shall be calculated (e.g. in the (R)AN, UPF, UE).
Every standardized 5QI (of GBR and Delay-critical GBR resource type) is associated with a default value for the Averaging window (specified in QoS characteristics Table 5.7.4.1). The averaging window may also be signalled together with a standardized 5QI to the (R)AN and UPF and if it is received, it shall be used instead of the default value.
The Averaging window may also be signalled together with a pre-configured 5QI to the (R)AN and if it is received, it shall be used instead of the pre-configured value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.3.7 Maximum Data Burst Volume | Each GBR QoS Flow with Delay-critical resource type shall be associated with a Maximum Data Burst Volume (MDBV).
MDBV denotes the largest amount of data that the 5G-AN is required to serve within a period of 5G-AN PDB.
Every standardized 5QI (of Delay-critical GBR resource type) is associated with a default value for the MDBV (specified in QoS characteristics Table 5.7.4.1). The MDBV may also be signalled together with a standardized 5QI to the (R)AN and if it is received, it shall be used instead of the default value.
The MDBV may also be signalled together with a pre-configured 5QI to the (R)AN and if it is received, it shall be used instead of the pre-configured value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.4 Standardized 5QI to QoS characteristics mapping | Standardized 5QI values are specified for services that are assumed to be frequently used and thus benefit from optimized signalling by using standardized QoS characteristics. Dynamically assigned 5QI values (which require a signalling of QoS characteristics as part of the QoS profile) can be used for services for which standardized 5QI values are not defined. The one-to-one mapping of standardized 5QI values to 5G QoS characteristics is specified in table 5.7.4-1.
Table 5.7.4-1: Standardized 5QI to QoS characteristics mapping
5QI
Value
Resource Type
Default Priority Level
Packet Delay Budget
(NOTE 3)
Packet Error
Rate
Default Maximum Data Burst Volume
(NOTE 2)
Default
Averaging Window
Example Services
1
GBR
20
100 ms
(NOTE 11,
NOTE 13)
10-2
N/A
2000 ms
Conversational Voice
2
(NOTE 1)
40
150 ms
(NOTE 11,
NOTE 13)
10-3
N/A
2000 ms
Conversational Video (Live Streaming)
3
30
50 ms
(NOTE 11,
NOTE 13)
10-3
N/A
2000 ms
Real Time Gaming, V2X messages (see TS 23.287 [121]).
Electricity distribution – medium voltage, Process automation monitoring
4
50
300 ms
(NOTE 11,
NOTE 13)
10-6
N/A
2000 ms
Non-Conversational Video (Buffered Streaming)
65
(NOTE 9,
NOTE 12)
7
75 ms
(NOTE 7, NOTE 8)
10-2
N/A
2000 ms
Mission Critical user plane Push To Talk voice (e.g. MCPTT)
66
(NOTE 12)
20
100 ms
(NOTE 10,
NOTE 13)
10-2
N/A
2000 ms
Non-Mission-Critical user plane Push To Talk voice
67
(NOTE 12)
15
100 ms
(NOTE 10,
NOTE 13)
10-3
N/A
2000 ms
Mission Critical Video user plane
75
(NOTE 14)
25
50 ms
(NOTE 13)
10-2
N/A
2000 ms
V2X messages (see TS 23.287 [121]).
A2X messages (see TS 23.256 [136])
71
56
150 ms (NOTE 11, NOTE 13, NOTE 15)
10-6
N/A
2000 ms
"Live" Uplink Streaming (e.g. TS 26.238 [76])
72
56
300 ms (NOTE 11, NOTE 13, NOTE 15)
10-4
N/A
2000 ms
"Live" Uplink Streaming (e.g. TS 26.238 [76])
73
56
300 ms (NOTE 11, NOTE 13, NOTE 15)
10-8
N/A
2000 ms
"Live" Uplink Streaming (e.g. TS 26.238 [76])
74
56
500 ms (NOTE 11, NOTE 15)
10-8
N/A
2000 ms
"Live" Uplink Streaming (e.g. TS 26.238 [76])
76
56
500 ms (NOTE 11, NOTE 13, NOTE 15)
10-4
N/A
2000 ms
"Live" Uplink Streaming (e.g. TS 26.238 [76])
5
Non-GBR
10
100 ms
(NOTE 10,
NOTE 13)
10-6
N/A
N/A
IMS Signalling
6
(NOTE 1)
60
300 ms
(NOTE 10,
NOTE 13)
10-6
N/A
N/A
Video (Buffered Streaming)
TCP-based (e.g. www, e-mail, chat, ftp, p2p file sharing, progressive video, etc.), AI/ML model download for image recognition (e.g. for model topology) (see TS 22.261 [2])
7
70
100 ms
(NOTE 10,
NOTE 13)
10-3
N/A
N/A
Voice,
Video (Live Streaming)
Interactive Gaming, AI/ML model download for image recognition (e.g. for model weight factors) (see TS 22.261 [2])
8
80
300 ms
(NOTE 10, NOTE 13)
10-6
N/A
N/A
Video (Buffered Streaming)
TCP-based (e.g. www, e-mail, chat, ftp, p2p file sharing, progressive
9
90
video, etc.)
10
90
1100ms
(NOTE 10,NOTE 13, NOTE 17,
NOTE 18)
10-6
N/A
N/A
Video (Buffered Streaming)
TCP-based (e.g. www, e-mail, chat, ftp, p2p file sharing, progressive video, etc.) and any service that can be used over satellite access type with these characteristics
69
(NOTE 9, NOTE 12)
5
60 ms
(NOTE 7, NOTE 8)
10-6
N/A
N/A
Mission Critical delay sensitive signalling (e.g. MC-PTT signalling)
70
(NOTE 12)
55
200 ms
(NOTE 7,
NOTE 10)
10-6
N/A
N/A
Mission Critical Data (e.g. example services are the same as 5QI 6/8/9)
79
65
50 ms
(NOTE 10,
NOTE 13)
10-2
N/A
N/A
V2X messages (see TS 23.287 [121])
80
68
10 ms
(NOTE 5,
NOTE 10)
10-6
N/A
N/A
Low Latency eMBB applications Augmented Reality
82
Delay-critical GBR
19
10 ms
(NOTE 4)
10-4
255 bytes
2000 ms
Discrete Automation (see TS 22.261 [2])
83
22
10 ms
(NOTE 4)
10-4
1354 bytes
(NOTE 3)
2000 ms
Discrete Automation (see TS 22.261 [2]);
V2X messages (UE - RSU Platooning, Advanced Driving: Cooperative Lane Change with low LoA. See TS 22.186 [111], TS 23.287 [121])
84
24
30 ms
(NOTE 6)
10-5
1354 bytes
(NOTE 3)
2000 ms
Intelligent transport systems (see TS 22.261 [2])
85
21
5 ms
(NOTE 5)
10-5
255 bytes
2000 ms
Electricity Distribution- high voltage (see TS 22.261 [2]).
V2X messages (Remote Driving. See TS 22.186 [111], NOTE 16, see TS 23.287 [121]).
Split AI/ML inference - DL Split AI/ML image recognition, (see TS 22.261 [2])
86
18
5 ms
(NOTE 5)
10-4
1354 bytes
2000 ms
V2X messages (Advanced Driving: Collision Avoidance, Platooning with high LoA. See TS 22.186 [111], TS 23.287 [121])
87
25
5 ms (NOTE 4)
10-3
500 bytes
2000 ms
Interactive Service - Motion tracking data, (see TS 22.261 [2])
88
25
10 ms (NOTE 4)
10-3
1125 bytes
2000 ms
Interactive Service - Motion tracking data, (see TS 22.261 [2]), split AI/ML inference - UL Split AI/ML image recognition, (see TS 22.261 [2])
89
25
15 ms (NOTE 4)
10-4
17000 bytes
2000 ms
Visual content for cloud/edge/split rendering (see TS 22.261 [2])
90
25
20 ms (NOTE 4)
10-4
63000 bytes
2000 ms
Visual content for cloud/edge/split rendering (see TS 22.261 [2])
NOTE 1: A packet which is delayed more than PDB is not counted as lost, thus not included in the PER.
NOTE 2: It is required that default MDBV is supported by a PLMN supporting the related 5QIs.
NOTE 3: The Maximum Transfer Unit (MTU) size considerations in clause 9.3 and Annex J are also applicable. IP fragmentation may have impacts to CN PDB and details are provided in clause 5.6.10.
NOTE 4: A static value for the CN PDB of 1 ms for the delay between a UPF terminating N6 and a 5G-AN should be subtracted from a given PDB to derive the packet delay budget that applies to the radio interface. When a dynamic CN PDB is used, see clause 5.7.3.4.
NOTE 5: A static value for the CN PDB of 2 ms for the delay between a UPF terminating N6 and a 5G-AN should be subtracted from a given PDB to derive the packet delay budget that applies to the radio interface. When a dynamic CN PDB is used, see clause 5.7.3.4.
NOTE 6: A static value for the CN PDB of 5 ms for the delay between a UPF terminating N6 and a 5G-AN should be subtracted from a given PDB to derive the packet delay budget that applies to the radio interface. When a dynamic CN PDB is used, see clause 5.7.3.4.
NOTE 7: For Mission Critical services, it may be assumed that the UPF terminating N6 is located "close" to the 5G_AN (roughly 10 ms) and is not normally used in a long distance, home routed roaming situation. Hence a static value for the CN PDB of 10 ms for the delay between a UPF terminating N6 and a 5G_AN should be subtracted from this PDB to derive the packet delay budget that applies to the radio interface.
NOTE 8: In RRC_IDLE, RRC_INACTIVE and RRC_CONNECTED mode, the PDB requirement for these 5QIs can be relaxed (but not to a value greater than 320 ms) for the first packet(s) in a downlink data or signalling burst in order to permit reasonable battery saving (DRX) techniques.
NOTE 9: It is expected that 5QI-65 and 5QI-69 are used together to provide Mission Critical Push to Talk service (e.g. 5QI-5 is not used for signalling). It is expected that the amount of traffic per UE will be similar or less compared to the IMS signalling.
NOTE 10: In RRC_IDLE, RRC_INACTIVE and RRC_CONNECTED mode, the PDB requirement for these 5QIs can be relaxed for the first packet(s) in a downlink data or signalling burst in order to permit battery saving (DRX) techniques.
NOTE 11: In RRC_IDLE and RRC_INACTIVE mode, the PDB requirement for these 5QIs can be relaxed for the first packet(s) in a downlink data or signalling burst in order to permit battery saving (DRX) techniques.
NOTE 12: This 5QI value can only be assigned upon request from the network side. The UE and any application running on the UE is not allowed to request this 5QI value.
NOTE 13: A static value for the CN PDB of 20 ms for the delay between a UPF terminating N6 and a 5G-AN should be subtracted from a given PDB to derive the packet delay budget that applies to the radio interface.
NOTE 14: This 5QI is only used for transmission of V2X messages as defined in TS 23.287 [121] and transmission of A2X messages as defined in TS 23.256 [136].
NOTE 15: For "live" uplink streaming (see TS 26.238 [76]), guidelines for PDB values of the different 5QIs correspond to the latency configurations defined in TR 26.939 [77]. In order to support higher latency reliable streaming services (above 500ms PDB), if different PDB and PER combinations are needed these configurations will have to use non-standardised 5QIs.
NOTE 16: These services are expected to need much larger MDBV values to be signalled to the RAN. Support for such larger MDBV values with low latency and high reliability is likely to require a suitable RAN configuration, for which, the simulation scenarios in TR 38.824 [112] may contain some guidance.
NOTE 17: The worst case one way propagation delay for GEO satellite is expected to be ~270ms,~21 ms for LEO at 1200km and 13 ms for LEO at 600km. The UL scheduling delay that needs to be added is also typically two way propagation delay e.g. ~540ms for GEO, ~42ms for LEO at 1200km and ~26 ms for LEO at 600km. Based on that, the 5G-AN Packet delay budget is not applicable for 5QIs that require 5G-AN PDB lower than the sum of these values when the specific types of satellite access are used (see TS 38.300 [27]). 5QI-10 can accommodate the worst case PDB for GEO satellite type.
NOTE 18: When UE is accessing the network via satellite access supporting regenerative payload, a static value for the CN PDB of ~290ms for GEO satellite, ~41 ms for LEO at 1200km and 33 ms for LEO at 600km for the delay between a UPF terminating N6 and a 5G-AN should be subtracted from a given PDB to derive the packet delay budget that applies to the radio interface.
NOTE: It is preferred that a value less than 64 is allocated for any new standardised 5QI of Non-GBR resource type. This is to allow for option 1 to be used as described in clause 5.7.1.3 (as the QFI is limited to less than 64). |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.5 Reflective QoS | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.5.1 General | Reflective QoS enables the UE to map UL User Plane traffic to QoS Flows without SMF provided QoS rules and it applies for IP PDU Session and Ethernet PDU Session. This is achieved by creating UE derived QoS rules in the UE based on the received DL traffic. It shall be possible to apply Reflective QoS and non-Reflective QoS concurrently within the same PDU Session.
For a UE supporting Reflective QoS functionality, the UE shall create a UE derived QoS rule for the uplink traffic based on the received DL traffic if Reflective QoS function is used by the 5GC for some traffic flows. The UE shall use the UE derived QoS rules to determine mapping of UL traffic to QoS Flows.
If the 3GPP UE supports Reflective QoS functionality, the UE should indicate support of Reflective QoS to the network (i.e. SMF) for every PDU Session. For PDU Sessions established in EPS and PDU Sessions transferred from EPS without N26 interface, the UE indicates Reflective QoS support using the PDU Session Establishment procedure. After the first inter-system change from EPS to 5GS for PDU Sessions established in EPS and transferred from EPS with N26 interface, the UE indicates Reflective QoS support using the PDU Session Modification procedure as described in clause 5.17.2.2.2. The UE as well as the network shall apply the information whether or not the UE indicated support of Reflective QoS throughout the lifetime of the PDU Session.
NOTE: The logic driving a supporting UE under exceptional circumstances to not indicate support of Reflective QoS for a PDU Session is implementation dependent.
Under exceptional circumstances, which are UE implementation dependent, the UE may decide to revoke previously indicated support for Reflective QoS using the PDU Session Modification procedure. In such a case, the UE shall delete all derived QoS rules for this PDU Session and the network shall stop any user plane enforcement actions related to Reflective QoS for this PDU Session. In addition, the network may provide signalled QoS rules for the SDFs for which Reflective QoS was used before. The UE shall not indicate support for Reflective QoS for this PDU Session for the remaining lifetime of the PDU Session.
If under the same exceptional circumstances mentioned above and while NAS level MM or SM congestion control timer is running, the UE needs to revoke a previously indicated support for Reflective QoS, the UE performs PDU Session Release procedure that is exempt from MM and SM congestion control as defined in clause 5.19.7. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.5.2 UE Derived QoS Rule | The UE derived QoS rule contains following parameters:
- One UL Packet Filter (in the Packet Filter Set as defined in clause 5.7.6);
- QFI;
- Precedence value (see clause 5.7.1.9).
Upon receiving DL packet, one UL Packet Filter derived from the received DL packet as described in this clause is used to identify a UE derived QoS rule within a PDU Session.
For PDU Session of IP type, the UL Packet Filter is derived based on the received DL packet as follows:
- When Protocol ID / Next Header is set to TCP or UDP, by using the source and destination IP addresses, source and destination port numbers and the Protocol ID / Next Header field itself.
- When Protocol ID / Next Header is set to UDP, if the received DL packet is UDP-encapsulated IPSec protected packet, by using the source and destination IP addresses, source and destination port numbers, the Security Parameter Index and the Protocol ID / Next Header field itself. In this case, if an uplink IPSec SA corresponding to a downlink IPSec SA of the SPI in the DL packet exists and the SPI of the uplink IPSec SA is known to the NAS layer, then the UL Packet Filter contains an SPI of the uplink IPSec SA.
- When Protocol ID / Next Header is set to ESP, by using the source and destination IP addresses, the Security Parameter Index and the Protocol ID / Next Header field itself. If the received DL packet is an IPSec protected packet and an uplink IPSec SA corresponding to a downlink IPSec SA of the SPI in the DL packet exists and the SPI of the uplink IPSec SA is known to the NAS layer, then the UL Packet Filter contains an SPI of the uplink IPSec SA.
NOTE 1: In this Release of the specification for PDU Sessions of IP type the use of Reflective QoS is restricted to service data flows for which Protocol ID / Next Header is set to TCP, UDP or ESP.
NOTE 2: The UE does not verify whether the downlink packets with RQI indication match the restrictions on Reflective QoS.
NOTE 3: How to determine the received DL packet is UDP-encapsulated IPSec protected packet is defined in RFC 3948 [138]. UDP encapsulation for ESP is used when a NAT is detected and there can be different Security Parameter Indexes within the same IP-tuples.
NOTE 4: Despite the indication of support for Reflective QoS, the UE might not be able to derive QoS Rules for ESP IPSec packets and UDP-encapsulated IPSec packets, if uplink IPSec SA corresponding to a downlink IPSec SA of the SPI in the DL packet exists and the SPI of the uplink IPSec SA is not known to the NAS layer.
For PDU Session of Ethernet type the UL Packet Filter is derived based on the received DL packet by using the source and destination MAC addresses, the Ethertype on received DL packet is used as Ethertype for UL packet. In the case of presence of IEEE Std 802.1Q [98], the VID and PCP in IEEE Std 802.1Q [98] header(s) of the received DL packet is also used as the VID and PCP field for the UL Packet Filter. When double IEEE Std 802.1Q [98] tagging is used, only the outer (S-TAG) is taken into account for the UL Packet Filter derivation.
NOTE 5: In this Release of the specification for PDU Sessions of Ethernet type the use of Reflective QoS is restricted to service data flows for which 802.1Q [98] tagging is used.
The QFI of the UE derived QoS rule is set to the value received in the DL packet.
When Reflective QoS is activated the precedence value for all UE derived QoS rules is set to a standardised value. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.5.3 Reflective QoS Control | Reflective QoS is controlled on per-packet basis by using the Reflective QoS Indication (RQI) in the encapsulation header on N3 (and N9) reference point together with the QFI and together with a Reflective QoS Timer (RQ Timer) value that is either signalled to the UE upon PDU Session Establishment (or upon PDU Session Modification as described in clause 5.17.2.2.2) or set to a default value. The RQ Timer value provided by the core network is at the granularity of PDU Session (the details are specified in TS 24.501 [47]).
When the 5GC determines that Reflective QoS has to be used for a specific SDF belonging to a QoS Flow, the SMF shall provide the RQA (Reflective QoS Attribute) within the QoS Flow's QoS profile to the NG-RAN on N2 reference point unless it has been done so before. When the RQA has been provided to the NG-RAN for a QoS Flow and the 5GC determines that the QoS Flow carries no more SDF for which Reflective QoS has to be used, the SMF should signal the removal of the RQA (Reflective QoS Attribute) from the QoS Flow's QoS profile to the NG-RAN on N2 reference point.
NOTE 1: The SMF could have a timer to delay the sending of the removal of the RQA. This would avoid signalling to the RAN in the case of new SDFs subject to Reflective QoS are bound to this QoS Flow in the meantime.
When the 5GC determines to use Reflective QoS for a specific SDF, the SMF shall ensure that the UPF applies the RQI marking (e.g. by setting the indication to use Reflective QoS in the QER associated with the DL PDR if not already set) for this SDF. The SMF shall also ensure that the uplink packets for this SDF can be received by the UPF from the QoS Flow to which the DL PDR of the SDF is associated with as specified in TS 29.244 [65], e.g. by generating a new UL PDR for this SDF for that QoS Flow and providing it to the UPF.
When the UPF is instructed by the SMF to apply RQI marking, the UPF shall set the RQI in the encapsulation header on the N3 (or N9) reference point for every DL packet corresponding to this SDF.
When an RQI is received by (R)AN in a DL packet on N3 reference point, the (R)AN shall indicate to the UE the QFI and the RQI of that DL packet.
Upon reception of a DL packet with RQI:
- if a UE derived QoS rule with a Packet Filter corresponding to the DL packet does not already exist,
- the UE shall create a new UE derived QoS rule with a Packet Filter corresponding to the DL packet (as described in clause 5.7.5.2); and
- the UE shall start, for this UE derived QoS rule, a timer set to the RQ Timer value.
- otherwise,
- the UE shall restart the timer associated to this UE derived QoS rule; and
- if the QFI associated with the downlink packet is different from the QFI associated with the UE derived QoS rule, the UE shall update this UE derived QoS rule with the new QFI.
NOTE 2: Non-3GPP ANs does not need N2 signalling to enable Reflective QoS. Non 3GPP accesses are expected to send transparently the QFI and RQI to the UE. If the UPF does not include the RQI, no UE derived QoS rule will be generated. If RQI is included to assist the UE to trigger an update of the UE derived QoS rule, the reception of PDU for a QFI restarts the RQ Timer.
Upon timer expiry associated with a UE derived QoS rule the UE deletes the corresponding UE derived QoS rule.
When the 5GC determines not to use Reflective QoS for a specific SDF any longer:
- The SMF shall ensure that the UPF stops applying RQI marking as specified in TS 29.244 [65] (e.g. by removing the indication to use Reflective QoS from the QER associated with the DL PDR) for this SDF.
- When the UPF receives this instruction to stop applying RQI marking, the UPF shall no longer set the RQI in the encapsulation header on the N3 (or N9) reference point DL packets corresponding to this SDF.
- The SMF shall also ensure that, after an operator configurable time, the uplink packets for this SDF will not be accepted by the UPF over the QoS Flow on which Reflective QoS was applied for this SDF as specified in TS 29.244 [65], e.g. by removing the UL PDR for this SDF from that QoS Flow.
NOTE 3: The operator configurable time has to be at least as long as the RQ Timer value to ensure that no UL packet would be dropped until the UE derived QoS rule is deleted by the UE.
When the 5GC determines to change the binding of the SDF while Reflective QoS is used for this SDF, the SMF shall ensure that the uplink packets for this SDF are accepted over the newly bound QoS Flow and, for an operator configurable time, over the previously bound QoS Flow. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.6 Packet Filter Set | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.6.1 General | The Packet Filter Set is used in the QoS rule and the PDR to identify one or more packet (IP or Ethernet) flow(s).
NOTE 1: A QoS Flow is characterised by PDR(s) and QoS rule(s) as described in clause 5.7.1.1.
NOTE 2: DL Packet Filter in a Packet Filter Set of a QoS rule may be needed by the UE e.g. for the purpose of IMS precondition.
The Packet Filter Set may contain one or more Packet Filter(s). Every Packet Filter is applicable for the DL direction, the UL direction or both directions.
NOTE 3: The Packet Filter in the Packet Filter Set of the default QoS rule that allows all UL traffic (also known as match-all Packet Filter) is described in TS 24.501 [47].
There are two types of Packet Filter Set, i.e. IP Packet Filter Set and Ethernet Packet Filter Set, corresponding to those PDU Session Types. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.6.2 IP Packet Filter Set | For IP PDU Session Type, the Packet Filter Set shall support Packet Filters based on at least any combination of:
- Source/destination IP address or IPv6 prefix.
- Source / destination port number.
- Protocol ID of the protocol above IP/Next header type.
- Type of Service (TOS) (IPv4) / Traffic class (IPv6) and Mask.
- Flow Label (IPv6).
- Security parameter index.
- Packet Filter direction.
NOTE 1: A value left unspecified in a Packet Filter matches any value of the corresponding information in a packet.
NOTE 2: An IP address or Prefix can be combined with a prefix mask.
NOTE 3: Port numbers can be specified as port ranges.
NOTE 4: Type of Service (IPv4)/Traffic class (IPv6) can be used to define packet filters for DSCP and ECN as described in RFC 3168 [193].
- (S)RTP Multiplexed Media Identification Information including a combination of at least one of the following:
- Synchronization Source (SSRC), as defined by IETF RFC 3550 [185];
- Payload Type (PT), as defined by IETF RFC 3550 [185];
- RTCP SDES item Media Identification (MID) and RTP SDES header extension for MID, as defined by IETF RFC 9143 [207];
- RTCP packet type. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.6.3 Ethernet Packet Filter Set | For Ethernet PDU Session Type, the Packet Filter Set shall support Packet Filters based on at least any combination of:
- Source/destination MAC address.
- Ethertype as defined in IEEE 802.3 [131].
- Customer-VLAN tag (C-TAG) and/or Service-VLAN tag (S-TAG) VID fields as defined in IEEE Std 802.1Q [98].
- Customer-VLAN tag (C-TAG) and/or Service-VLAN tag (S-TAG) PCP/DEI fields as defined in IEEE Std 802.1Q [98].
- IP Packet Filter Set, in the case that Ethertype indicates IPv4/IPv6 payload.
- Packet Filter direction.
NOTE 1: The MAC address can be specified as address ranges.
NOTE 2: A value left unspecified in a Packet Filter matches any value of the corresponding information in a packet. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.7 PDU Set QoS Parameters | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.7.1 General | PDU Set QoS Parameters are used to support PDU Set based QoS handling in the 5G-AN.
The following PDU Set QoS Parameters are specified:
1. PDU Set Delay Budget (PSDB).
2. PDU Set Error Rate (PSER).
3. PDU Set Integrated Handling Information (PSIHI).
At least one of the following shall be sent to the NG-RAN/N3IWF/TNGF/W-AGF to enable PDU Set based handling: 1) a PSIHI and/or 2) both PSDB and PSER. For a given QoS Flow, the values of PSDB, PSER and PSIHI can be different for UL and DL.
The QoS Profile may include the PDU Set QoS Parameters described in this clause (see clause 5.7.1.2) for UL and/or DL direction. The PCF determines the PDU Set QoS Parameters based on information provided by AF and/or local configuration. The PDU Set QoS parameters are sent to the SMF as part of PCC rule. The SMF sends them to 5G-AN as part of the QoS Profile.
If the 5G-AN receives PDU Set QoS Parameters, it enables the PDU Set based QoS handling and applies PDU Set QoS Parameters as described in TS 38.300 [27], TS 38.413 [34] and TS 38.331 [51]. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.7.2 PDU Set Delay Budget | The PDU Set Delay Budget (PSDB) defines an upper bound for the delay that a PDU Set may experience for the transfer between the UE and the N6 termination point at the UPF, i.e. the duration between the reception time of the first PDU (at the N6 termination point for DL or the UE for UL) and the time when all PDUs of a PDU Set have been successfully received (at the UE for DL or N6 termination point for UL). The DL PSDB applies to the DL PDU Set received by the PSA UPF over the N6 interface and the UL PSDB applies to the UL PDU Set sent by the UE.
NOTE 1: To enable support for PSDB, it is required that a maximum inter arrival time between the first received PDU and the last received PDU of a PDU Set complies with SLA. This maximum inter arrival time does not exceed PSDB. 5G-AN behaviour when the SLA is not fulfilled is out of scope of this specification.
A QoS Flow is associated with at most one PSDB value per direction. PSDB is an optional parameter that may be provided by the PCF. The provided PSDB can be used by the 5G-AN to support the configuration of scheduling and link layer functions.
When the PSDB is available, the UL and/or DL PSDB supersedes the PDB in the respective direction for the given QoS Flow in the 5G-AN.
NOTE 2: See clause 5.7.7.1 for dependency between PDU Set QoS parameters.
The AN PSDB is derived at 5G-AN by subtracting CN PDB (as described in clause 5.7.3.4) from the PSDB. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.7.3 PDU Set Error Rate | The PDU Set Error Rate (PSER) defines an upper bound for the rate of PDU Sets that have been processed by the sender of a link layer protocol (e.g. RLC in RAN of a 3GPP access) but that are not successfully delivered by the corresponding receiver to the upper layer (e.g. PDCP in RAN of a 3GPP access). Thus, the PSER defines an upper bound for a rate of non-congestion related PDU Set losses. The purpose of the PSER is to allow for appropriate link layer protocol configurations (e.g. RLC and HARQ in RAN of a 3GPP access).
NOTE 1: In this Release, a PDU Set is considered as successfully delivered only when all PDUs of a PDU Set are delivered successfully.
NOTE 2: How 5G-AN enforces PSER is up to 5G-AN implementation.
A QoS Flow is associated with at most one PSER value per direction. PSER is an optional parameter. If the PSER is available, the UL and/or DL PSER supersedes the PER in the respective direction in the 5G-AN.
NOTE 3: See clause 5.7.7.1 for dependency between PDU Set QoS parameters. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.7.7.4 PDU Set Integrated Handling Information | The PDU Set Integrated Handling Information (PSIHI) indicates whether all PDUs of the PDU Set are needed for the usage of the PDU Set by the application layer in the receiver side. PSIHI is an optional parameter. A QoS Flow is associated with at most one PSIHI value per direction. |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.8 User Plane Management | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.8.1 General | User Plane Function(s) handle the user plane path of PDU Sessions. 3GPP specifications support deployments with a single UPF or multiple UPFs for a given PDU Session. UPF selection is performed by SMF. The details of UPF selection is described in clause 6.3.3. The number of UPFs supported for a PDU Session is unrestricted.
For an IPv4 type PDU Session or an IPv6 type PDU Session without multi-homing or an IPv4v6 type PDU Session, when multiple PDU Session Anchors are used (due to UL CL being inserted), only one IPv4 address and/or IPv6 prefix is allocated for the PDU Session. For an IPv6 multi-homed PDU Session there are multiple IPv6 prefixes allocated for the PDU Session as described in clause 5.6.4.3.
If the SMF had requested the UPF to proxy ARP or IPv6 Neighbour Solicitation for an Ethernet DNN, the UPF should respond to the ARP or IPv6 Neighbour Solicitation Request, itself.
Deployments with one single UPF used to serve a PDU Session do not apply to the Home Routed case and may not apply to the cases described in clause 5.6.4.
Deployments where a UPF is controlled either by a single SMF or multiple SMFs (for different PDU Sessions) are supported.
UPF traffic detection capabilities may be used by the SMF in order to control at least following features of the UPF:
- Traffic detection (e.g. classifying traffic of IP type, Ethernet type, or unstructured type)
- Traffic reporting (e.g. allowing SMF support for charging).
- QoS enforcement (The corresponding requirements are defined in clause 5.7).
- Traffic routing (e.g. as defined in clause 5.6.4. for UL CL or IPv6 multi-homing). |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.8.2 Functional Description | |
fbecc7f0dcf9784c6066646052ab0c0e | 23.501 | 5.8.2.1 General | This clause contains detailed functional descriptions for some of the functions provided by the UPF. It is described how the SMF instructs it's corresponding UP function and which control parameters are used. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.