hash stringlengths 32 32 | doc_id stringlengths 5 12 | section stringlengths 5 1.47k | content stringlengths 0 6.67M |
|---|---|---|---|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9 Solution #9: Support of SMS to Emergency Centre over NAS
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.0 High level principles
| This clause describes support for the emergency Short Message Service (SMS) over NAS. This solution is based on the architectures for SMS over NAS (see TS 23.272 [7], TS 23.401 [5], TS 23.501 [2], TS 23.502 [3]) and the architectural principles of Emergency Services (see TS 23.167 [9]).
SMS to Emergency Centre over NAS in this context is SMS over NAS using an emergency number as destination address. A key requirement for this service is that the emergency Short Message (SM) must be routed to a PSAP in the country where the UE initiating the emergency SM is located. This implies that SMS to Emergency Centre for roaming subscribers cannot be home routed while regular SMS is typically home routed.
This solution applies to UE detectable emergency numbers.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1 Description
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1.1 General
| This clause lists the additional functionality that is required to support SMS to Emergency Centre over NAS for roaming and non-roaming users.
High level description and signalling flows are described in TS 23.272 [7] Annex C for EPS and TS 23.502 [3] for 5GS. SMS to Emergency Centre over NAS in 5GS is based on the architecture defined in clause 4.4.2. It can be supported over 3GPP and non-3GPP access.
There is no capability negotiation between the UE and AMF. The UE that supports SMS to Emergency Centre over NAS shall replace the SMSC Address with the country code and the detected emergency number and the detected emergency number as destination address as defined in TS 23.167 [9] for the UE functionality.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1.2 Service assumptions
| The following service assumptions are considered for supporting SMS to Emergency Centre over NAS:
- The UE that supports SMS to Emergency Centre over NAS and detects SMS to Emergency Centre over NAS based on used emergency numbers. Provisioning of emergency numbers and associated types as specified in clause 4.1 bullets 2a and 2b) of TS 23.167 [9].
- SM messages from the PSAP to the UE are routed like any other SM.
- Serving network can support SMS to Emergency Centre over NAS for own subscribers and inbound roamers.
- SMS Centre in the serving network supports SMS to Emergency Centre over NAS.
- There is no indication from the serving PLMN to the UE that SMS to Emergency Centre in supported. After the UE sends SMS to Emergency Centre over NAS it can get informed about the success or failure based on the existing SMS delivery notification using existing procedures.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1.3 How the solution solves the key issues
| How does the solution address the study item
- How to identify a Short Message to Emergency Response Centre based on the use of emergency numbers.
- This is based on the existing routing of SMS and the emergency number used by the UE which is provisioned with emergency numbers and associated types. PLMNs already support detection of emergency numbers and the SMSF or the MME selects the Emergency Response Centre based on the emergency number
- Address aspects when serving PLMN does not support this feature, but UE supports it.
- Serving PLMNs are assumed to support emergency services to configured emergency numbers and thus support SMS to Emergency Centre over NAS and the ability to identify a Short Message to Emergency Response Centre.
- Whether and how to identify the type of emergency service (police, ambulance, fire brigade, etc.) using the same identification mechanism as for emergency calls.
- The UE uses the provisioned emergency numbers which identify the type of service and the user selects e.g. from UI of the phone with mechanisms out of scope of 3GPP.
- How to route a Short Message to Emergency Response Centre (i.e. PSAP) serving the UE's location, according to the local regulations (in roaming case, the regulations applicable to the UE's location).
- This is based on existing mechanisms for routing of SMS and in addition the SMSF or MME uses the UE location and the emergency number to select the PSAP.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1.4 UE Functionality
| The UE that supports SMS to Emergency Centre over NAS shall support the functionality listed below (which needs to be added to TS 23.167 [9]):
- Identify the need to send SMS to Emergency Centre over NAS based on the use of emergency numbers.
- Performs domain selection for SMS to Emergency Centre over NAS as described in TS 23.221 [12].
- Replaces the SC Address that is configured with country code of the UEs serving PLMN and the emergency number e.g. +32112.
NOTE 1: If UE cannot determine country from serving PLMN, UE can use methods out of scope of the present solution to determine the county.
NOTE 2: After the UE sends SMS to Emergency Centre over NAS it can get informed about the success or failure based on the SMS delivery notification using existing procedures.
Editor's note: Whether the UE needs to replace the SC Address or an alternative solution with no SC Address replacement is preferred needs to be determined during evaluation phase.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.1.5 Domain Selection for UE originating SMS
| To allow for appropriate domain selection for SMS delivery, the Domain Selection for UEs originating SMS in TS 23.221 [12] needs to be augmented with the following:
- If the UE does not support SMS to Emergency Centre over IMS and if the UE supports SMS to Emergency Centre over NAS and UE detects that the SM destination address is an emergency number as defined in TS 23.167 [9], the UE shall attempt to deliver the SMS over NAS signalling.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.2 Procedures
| The MO SMS over NAS in CM-IDLE (baseline) procedure in clause 4.13.3.3 of TS 23.502 [3] is modified with the following additions (in bold underline).
Figure 6.9.2-1: MO SMS over NAS
1. The UE performs domain selection for UE originating SMS as defined in clause 5.16.3.8 of TS 23.501 [2] if SMS delivery via non 3GPP access is allowed and possible. If an UE under CM-IDLE state is going to send uplink SMS message, then UE and network perform the UE Triggered Service Request procedure firstly as defined in clause 4.2.3.2 to establish a NAS signalling connection to AMF.
2a. The UE builds the SMS message to be sent as defined in TS 23.040 [13] (i.e. the SMS message consists of CP-DATA/RP-DATA/TPDU/SMS-SUBMIT parts). The SMS message is encapsulated in an NAS message with an indication indicating that the NAS message is for SMS transporting. The UE send the NAS message to the AMF. For SMS to Emergency Centre the UE follows the procedures in TS 23.167 [9].
2b. The AMF forwards the SMS message and SUPI to the SMSF serving the UE over N20 message by invoking Nsmsf_SMService_UplinkSMS service operation. In order to permit the SMSF to create an accurate charging record, the AMF adds the IMEISV, the current UE Location Information (ULI) of the UE as defined in clause 5.6.2 of TS 23.501 [2] and if the UE has sent the SMS via 3GPP access, the local time zone.
When the AMF determines that the UE has the MPS for Messaging indication set (enabled) in the UE context, the AMF includes a Message Priority header to indicate priority information. Other NFs relay the priority information by including the Message Priority header in service-based interfaces, as specified in TS 29.500 [14].
Editor's note: Whether there are any implications related to MPS for Messaging as defined in TS 23.501 [2] is FFS.
2c. The SMSF invokes Namf_Communication_N1N2MessageTransfer service operation to forward SMS ack message to AMF.
2d. The AMF forwards the SMS ack message from the SMSF to the UE using downlink unit data message.
3-5. The SMSF checks the SMS management subscription data. If SMS delivery is allowed, the procedure defined in TS 23.040 [13] or TS 23.540 [15] applies. The SMSF sets proper DRMP value based on received Message Priority header in step 2. If the SMSF supports SMS to Emergency Centre and detects from the SMSC Address and the destination address contained in the SM the SMSF that the SM is for emergency, it sends the SM to the local SMS-GMSC/IWMSC for emergency.
6a-6b. The SMSF forwards the submit report to AMF by invoking Namf_Communication_N1N2MessageTransfer service operation which is forwarded to UE via Downlink NAS transport. If the SMSF knows the submit report is the last message to be transferred for UE, the SMSF shall include a last message indication in the Namf_Communication_N1N2MessageTransfer service operation so that the AMF knows no more SMS data is to be forwarded to UE.
NOTE: The behaviour of AMF based on the "last message indication" is implementation specific.
If the UE has more than one SMS message to send, the AMF and SMSF forwards SMS /SMS ack/submit report the same way as described in step 2a-6b.
6c-6d. When no more SMS is to be sent, UE returns a CP-ack as defined in TS 23.040 [13] to SMSF. The AMF forwards the SMS ack message by invoking Nsmsf_SMService_UplinkSMS service operation to SMSF.
Editor's note: Whether this procedure can be reused for EPS by replacing AMF+SMSF by MME is FFS.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.9.3 Impacts to Services, Entities and Interfaces
| UE:
- detects that the SM destination address is an emergency number based on existing mechanisms and replaces the SC Address that is configured with country code of the UEs serving PLMN and the emergency number e.g. +32112.
SMSF/MME:
- process all MO-SMS and detects from the SMSC Address and the destination address contained in the SM the SMSF that the SM is for emergency, it sends the SM to the local SMS-GMSC/IWMSC for emergency.
- is configured to recognize emergency numbers of other countries for which inbound roamers are allowed to send SMS to Emergency Centre.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10 Solution #10: SMS over NAS in 4G/5G to Support SMS Delivery to the Emergency Response Centre
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.0 High level principles
| To support SMS delivery to Emergency Response Centres via NAS, the following high-level solution principles apply for both EPS and 5GS:
- The UE may include an explicit indication towards MME/AMF to indicate the MO data containing SMS to Emergency Response Centre (EC).
- For cases where the UE cannot detect the SMS to EC, the message is transmitted as a normal SMS, with identification of SMS to EC performed by the network.
- Upon detection of a SMS to EC, the MME/AMF may trigger UE location retrieval.
- Location information may be provided together with the SMS towards the SMSC/SMS Router for delivery to the Emergency Response Centre. The functionalities of SMS Router are specified in TS 23.040 [13], TS 23.204 [8] and TS 23.272 [7].
- The MME in EPS or SMSF in 5GS forwards the SMS TPDU to the appropriate SMS-GMSC/SMS-IWMSC/SMS Router for onward routing.
- The SMSC, based on user data header information in TPDU described in Solution 1, forwards the SMS to the operator's dedicated emergency service network or directly to the Emergency Response Centre, in accordance with local regulatory requirements.
These principles apply to both roaming and non-roaming scenarios. Callback from the Emergency Response Centre to the UE is supported by reusing the regular SMS delivery service. The SMS from the Emergency Response Centre is delivered via the standard SMS path through the serving PLMN's SMSC.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.1 Description
| In EPS, the UE delivers the SMS to EC using SMS over NAS. If the UE is in idle mode and detects the intent to send an SMS to EC, it initiates the UE-triggered Service Request procedure. If both the UE and the MME support Control Plane (CP) CIoT EPS Optimisation for SMS transfer, the UE initiates a Control Plane Service Request (CPSR).
When CP CIoT EPS Optimisation is used, the UE encapsulates the SMS-SUBMIT TPDU in a CPSR message towards the MME. Otherwise, the SMS to EC is encapsulated within the Uplink NAS Transport message. The UE may include an explicit SMS to EC indicator in the NAS message to inform the MME that the payload contains a SMS to EC in the SMS TPDU.
If the SMS is identified as a SMS to EC, the UE may include an emergency indication. The MME may use this indication to trigger UE location retrieval using EPC Network Induced Location Request (EPC-NI-LR) as specified in clause 9.1.17 of TS 23.271 [18] and/or bypass applicable barring checks. The MME then forwards the SMS TPDU, along with the obtained location information, to the SMS-GMSC/SMS-IWMSC/SMS Router over the SGd interface using the MO-Forward-Short-Message-Request procedure. The SMSC subsequently delivers the SMS to the Emergency Response Centre in accordance with local regulatory requirements.
In 5GS, the UE delivers the SMS to EC via an Uplink NAS Transport message towards the AMF. When the SMS is identified as a SMS to EC, the UE includes an indication in the NAS message. The AMF may use this indication to bypass barring checks and/or to trigger UE location retrieval using the 5GC Network Induced Location Request (5GC-NI-LR) as specified in clause 6.10 of TS 23.273 [17].
The AMF forwards the SMS to the serving SMSF using the Nsmsf_SMService_UplinkSMS operation. The SMSF examines the RP-DA, and forwards it to the appropriate SMS-GMSC/SMS-IWMSC/SMS Router, which delivers the message to the SMSC. The SMSC, based on information contained in SMS user data header, transfers the SMS to the Emergency Response Centre as required by local regulation.
Callback from the Emergency Response Centre to the UE is supported using the regular SMS delivery service. The Emergency Response Centre sends the SMS via the operator's SMSC and the SMS is delivered to the UE over the standard downlink SMS path.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.2 Procedures
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.2.1 SMS in MME for 4G
| Figure 6.10.2.1-1 shows the procedures for SMS delivery to the Emergency Response Centre via SMS over NAS using the SMS in MME architecture option as described in Annex C of TS 23.272 [7].
Figure 6.10.2.1-1: Procedures for Emergency SMS Delivery via SMS over NAS in EPC
0. The UE attaches to the network using either Initial attach or Emergency Attach.
1. If the UE is in idle mode and detects the intent to send a SMS to EC, it initiates the UE-triggered Service Request procedure. If both the UE and the MME support Control Plane (CP) CIoT EPS Optimisation for SMS transfer, the UE initiates a Control Plane Service Request (CPSR).
When CP CIoT EPS Optimisation is used, the CPSR message carries the SMS-SUBMIT TPDU directly. Otherwise, in step 1a, the SMS is encapsulated within the Uplink NAS Transport message.
The UE may include an explicit SMS to EC indicator in the NAS message to inform the MME that the payload contains an emergency message in the SMS TPDU.
2. If the UE does not know its own location, the UE may retrieve its location information from the network. Or based on the indication described in step 1, the MME may trigger UE location retrieval using EPC Network Induced Location Request (EPC-NI-LR) in step 2a as specified in clause 9.1.17 of TS 23.271 [18] and/or bypass applicable barring checks.
3. The MME selects the target SMS-GMSC/SMS-IWMSC/SMS Router and forwards the SMS TPDU along with the location information using the MO-Forward-Short-Message-Request (OFR) message over the SGd interface, as specified in TS 29.338 [16].
4. The SMS-GMSC/SMS-IWMSC/SMS Router, based on the RP-DA to route the message to the appropriate SMSC.
5. The SMSC, based on the SMS-TPDU user data header information as described in Solution 1, forwards the message to the operator's dedicated emergency service network. For example:
- Advanced Mobile Location (AML) over NAS/SGd, defined in GSMA NG.119 [11], allows the visited SMSC to send the SMS directly to the PSAP or Emergency Call Centre network.
- Text-to-9-1-1, as specified in J-STD-110 [10], enables the SMSC to forward the SMS via the SMPP protocol to the Text Control Centre (TCC), which functions as a gateway to the PSAP.
6. The Delivery Report is sent to UE.
Editor's note: Whether this procedure can be reused for EPS deployment in SGs option is FFS.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.2.2 SMS over NAS for 5G
| Figure 6.10.2.2-1 shows the procedures for SMS delivery to the Emergency Response Centre using SMS over NAS for 5G.
Figure 6.10.2.2-1: Procedures for Emergency SMS Delivery via SMS over NAS in 5G
0. The UE follows the procedures specified in clause 4.13.3 of TS 23.502 [3] for registration with SMS over NAS support. During the Registration procedure in 5GS, the UE includes an "SMS supported" indication in the Registration Request to indicate its capability for SMS over NAS transport.
1. If the UE is in CM-IDLE state and intends to send an uplink SMS message, the UE and the network perform the UE-triggered Service Request procedure as defined in clause 4.2.3.2 of TS 23.502 [3] to establish a NAS signalling connection to the AMF.
If the UE is in CM-CONNECTED state, this step is skipped and the procedure continues with step 2.
2. The UE builds the SMS message.
For a SMS to EC, the UE encapsulates the SMS in an Uplink NAS Transport message and includes a SMS to EC indication. The Uplink NAS Transport message is then sent to the AMF.
For a SMS to EC that is not detectable by the UE (e.g. based on number or configuration), the SMS is encapsulated in an Uplink NAS Transport message without the emergency indication and sent to the AMF.
3. If the UE does not know its own location, the UE may retrieve its location information from the network. Or based on the SMS to EC indication, the AMF may trigger UE location retrieval, using the 5GC Network Induced Location Request (5GC-NI-LR) in step 3a, as specified in clause 6.10 of TS 23.273 [17] and/or bypass applicable barring checks.
4. The AMF forwards the SMS to the appropriate SMSF serving emergency SMS by invoking the Nsmsf_SMService_UplinkSMS service operation.
5. The SMSF inspects the user data header contained in TPDU.
If the message is identified as a SMS to EC, the SMSF forwards it to the appropriate SMS-GMSC/SMS-IWMSC/SMS Router.
The SMSF then returns the SMS-Submit Report to the AMF using the Namf_Communication_N1N2MessageTransfer service operation. The AMF delivers the report to the UE via Downlink NAS Transport as described in clause 4.13.3.3 of TS 23.502 [3].
6. The SMS-GMSC/SMS-IWMSC/SMS Router, based on the RP-DA route the message to the appropriate SMSC.
7. The SMSC, based on the user data header information contained in TPDU, forwards the message to the operator's dedicated emergency service network. For example:
- Advanced Mobile Location (AML) over NAS/SGd, defined in GSMA NG.119 [11], enables the visited SMSC to deliver the SMS directly to the PSAP or Emergency Call Centre network.
- Text-to-9-1-1, as specified in J-STD-110 [10], enables the SMSC to forward the message via the SMPP protocol to the Text Control Centre (TCC), which functions as a gateway to the PSAP.
8. The Submit Report is sent to UE.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.10.3 Impacts to Services, Entities and Interfaces
| MME:
- MME uses SMS to EC indication to trigger location request.
AMF:
- AMF uses SMS to EC indication to trigger location request.
UE:
- UE indicates Emergency SMS indication towards MME/AMF.
SMSC:
- SMSC uses user data header information contained in TPDU to forward the message to a proper emergency service network.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.11 Solution #11: Emergency SMS to EC over NAS
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.11.0 High level principles
| The following principles are applied to this solution:
- The UE is normally registered with the network and emergency numbers are provisioned, as specified in TS 23.167 [9].
- SMS to EC via NAS is supported by both the UE and the network.
- The UE detects a MO SM addressed to an EC.
- The UE sending a SM to EC may send early indication of emergency communication via NAS.
- The SMSF and AMF send an acknowledgement to the UE.
- The SMSF forwards the report to the UE.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.11.1 Description
| Editor's note: EPS procedure is FFS.
This solution provides and end-to-end procedure to support the SMS to Emergency Response Centre via NAS.
After having registered to the network, the UE is authorised for SMS to EC. Depending on location and local regulation, it may also be provisioned with emergency numbers and corresponding service types (details will be addressed by solutions to KI#1).
Upon detecting an originating emergency SM, the UE sends an UL NAS transport message to the AMF containing the SMS body addressed to an EC. The AMF forwards it to the SMSF that requests the NRF and UDM for emergency SM service.
When granted service for emergency SMS, the SMSF forwards the emergency SM to the SMS-IWMSC or SMSC that sends it the appropriate EC and then the SMSF sends an ACK back to the UE.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.11.2 Procedures
| Editor's note: EPS procedure is FFS.
Editor's note: The deviation for the procedure of the MO SMS over NAS Figure 13.3.3-1 in TS 23.502 [3] is FFS.
The below figure shows the procedure to send an SMS to EC over the 5GS.
Figure 6.11.2-1: Procedure for SMS over NAS
0.0 Emergency number/information provisioning to the UE as specified in TS 23.167 [9].
0.1 UE originates a SM to EC.
1. The UE performs domain selection for UE originating SM as defined in clause 5.16.3.8 of TS 23.501 [2]
2a. The UE builds the SM to be sent to EC, as defined in TS 23.040 [13] (e.g. the SMS message consists of CP-DATA/RP-DATA/TPDU/SMS-SUBMIT parts). The SM is encapsulated in a NAS message with an indication indicating that the NAS message is for SMS to EC.
2b. In the 5GC, the AMF determines that the received message contains SMS to EC. The AMF forwards the SMS message/body and SUPI to the SMSF serving the corresponding UE2c. in the 5GC, the SMSF invokes Namf_Communication_N1N2MessageTransfer service operation to forward SMS ack message to AMF.
2d. The AMF forwards the SMS ack message from the SMSF to the UE using downlink unit data message via the RAN node.
3. The SMSF sends a Niwmsc_SMService_MoForwardSm service request to the URI of the serving SMS-IWMSC. The payload body of the request shall contain the SM record to be sent, the Service/emergency Centre address.
4. The SMS is delivered to EC from SMS-IWMSC or SMSC (for emergency services).
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.11.3 Impacts to Services, Entities and Interfaces
| The solution has the following impacts:
UE:
- Need to support indication to serving PLMN support for SMS to EC during Attach/Registration.
- Need to support a new NAS payload type for SMS to EC when initiating MO SMS to EC.
AMF:
- Need to support indication to UE of support for SMS to EC during Attach/Registration.
- Need to support a new NAS payload type for SMS to EC.
SMSF:
- Support routing to local SMS-IWMSC or SMSC for SMS to EC.
SMS-IWMSC or SMSC:
- Supports routing to EC.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 7 Interim Agreements
| Editor's note: This clause will capture interim agreements derived from solutions.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 8 Conclusions
| Editor's note: This clause will list conclusions that have been agreed during the course of the study item activities.
Annex A:
Change history
Change history
Date
Meeting
TDoc
CR
Rev
Cat
Subject/Comment
New version
2025-08
SA2#170
S2-2507491
-
-
-
Proposed skeleton agreed for FS_SMS2EC_ARC at SA2#170
0.0.0
2025-09
SA2#170
-
-
-
-
Implementing the following approved pCRs: S2-2507492, S2-2507594, S2-2507595, S2-2507596, S2-2507644
0.1.0
2025-10
SA2#171
-
-
-
-
Implementing the following approved pCRs: S2-2509350, S2-2509353, S2-2509585, S2-2509586, S2-2509588, S2-2509589, S2-2509590, S2-2509615, S2-2509616, S2-2509617, S2-2509619, S2-2509620
0.2.0
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1 Solution #1: Support of SMS to EC indication and SMS user data header for SMS to EC
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.0 High level principles
| The following high-level principles are proposed for the support of SMS to Emergency Centre (EC) service:
- A header is introduced within the SMS-SUBMIT TPDU, specifically within the TP-UDH. The header may include emergency-related information such as the requested emergency service type, callback identifiers, or other parameters required for accurate routing to a PSAP.
- When the UE determines that a message is a SMS towards EC message, the UE composes the SMS-SUBMIT TPDU including the header with emergency-related information and submits it using a Service Centre (SC) address configured for handling SMS to EC as outlined in GSMA NG.119 [11]. The SC address is carried in the SMS-SUBMIT RPDU (Relay Protocol Data Unit).
- The UE may also indicate towards the network at NAS layer that the MO data transport contains a SMS to EC.
- The serving network forwards the SMS to the SMSC. The SMSC applies emergency-specific processing and delivers the message to the appropriate PSAP, supporting use cases such as Text-to-9-1-1 (see ATIS J-STD-110 [10]).
- For cases where the UE does not detect that the SMS is a SMS to EC, the message is composed as a normal SMS without header with emergency-related information. The SMSC may inspect the Destination Address (TP-DA) of the SMS. If the TP-DA matches an entry in the network's provisioned emergency-number list, the SMS is classified as a SMS to EC. The network shall then internally flag the SMS as a SMS to EC and ensure that it is routed to the correct PSAP in accordance with applicable regulatory requirements.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.1 Description
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.1.1 UE Detected and Non UE Detectable SMS to EC
| To support reliable detection and routing of SMS to EC, the UE may include a header within the TP-UDH of the SMS-SUBMIT TPDU. The header may carry critical information such as a callback number, emergency service type and location information e.g. pre-registered civic address for the purpose of emergency, GPS/device-based location, or network-based location. Provision of such information at the point of message origination enables the network to more reliably determine whether a MO SMS requires emergency handling and to ensure accurate routing to the appropriate Public Safety Answering Point (PSAP). This mechanism reduces the risk of misrouted or delayed SMS to EC and thereby improves user safety.
The Destination Address TP-DA shall correspond to an emergency short code (e.g. 112, 911). The header in TP-UDH may contain the following emergency-related information:
- indication that the TP-UD contains a message to EC;
- callback number, if available;
- location information (e.g. pre-registered civic address, GPS/device-based location, or network-based location), if available;
- emergency service type (e.g. police, ambulance, fire).
The serving network may download additional emergency numbers to the UE to enable the UE to detect a SMS to EC.
In the case of non-UE detectable SMS to EC, the message will be handled as a normal SMS. The SMSC shall inspect the TP-DA of the SMS-SUBMIT TPDU. If the TP-DA matches an entry in the provisioned emergency-number list, the message shall be flagged as emergency traffic and routed to the appropriate PSAP in accordance with regulatory requirements.
The SMSC or SMS Router shall inspect the SMS-SUBMIT TPDU. If the header with emergency-related information is present in the TP-UDH, the network shall extract the included information (e.g. callback number, emergency service type and location information) and classify the message as a SMS to EC. The network may query the location retrieval functionalities in the network to obtain authoritative location information, apply local routing policies and ensure that the message is delivered to the correct PSAP in the serving country or region.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.2 Procedures
| |
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.2.1 Roaming Handling
| For emergency messages initiated by a roaming user, the principle of Local Breakout is maintained, prioritizing service from the VPLMN based on local regulations. If the UE does not know its own location, the UE may retrieve its location information from the VPLMN, e.g. LRF lookup and the VPLMN routes the message directly to the local PSAP. The user data header contained in the TPDU maybe used to choose the correct service-type PSAP.
Editor's note: It is FFS if VPLMN does not support the required capability to route MO SMS to the local PSAP.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.1.3 Impacts to Services, Entities and Interfaces
| UE:
- Support the inclusion of a header with emergency-related information in the TP-UDH of SMS-SUBMIT when originating SMS to EC.
- Indicate towards the NAS layer that the MO data transport contains SMS to EC.
SMSC/SMS Router:
- Inspect SMS-SUBMIT TPDU for presence of the header with emergency-related information in TP-UDH.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.2.2.1.1 Registration procedure for SMS over NAS
| The following procedures are based on clause 4.13.3 of TS 23.502 [3] with the following additions (in bold underline).
Figure 6.2.2.1.1-1: Based on TS 23.502 [3], Figure 4.13.3.1-1: Registration procedure supporting SMS over NAS
1. During Registration procedure in 5GS defined in Figure 4.2.2.2.2-1, to enable SMS over NAS transporting, the UE includes an "SMS supported" indication in Registration Request in step 1-3 indicating the UE's capability for SMS over NAS transport. The "SMS supported" indication indicates whether the UE supports SMS delivery over NAS.
2. Step 4 to step 14 of the Registration procedure in Figure 4.2.2.2.2-1 are performed. The AMF may retrieve the SMS Subscription data and UE Context in SMSF data using Nudm_SDM_Get. This requires that UDM may retrieve this information from UDR by Nudr_DM_Query. The UDM includes the SMSF information in the Nudm_SDM_Get response message if the stored SMSF belongs to the same PLMN of the AMF. After a successful response is received and if SMS service is allowed, the AMF subscribes to be notified using Nudm_SDM_Subscribe when the SMS Subscription data is modified and UDM may subscribe to UDR by Nudr_DM_Subscribe. For an MPS-subscribed UE, the subscription data may include the MPS priority for Messaging indication. If the AMF receives the MPS for Messaging indication from UDM, the AMF stores the MPS for Messaging indication in the UE context.
The AMF can also receive UE context information containing SMSF Information from old AMF. When AMF re-allocation happens during the Registration procedure, the old AMF transfers SMSF Information to the new AMF as part of UE context in step 5 of Figure 4.2.2.2.2-1.
NOTE 1: The AMF can, instead of the Nudm_SDM_Get service operation, use the Nudm_SDM_Subscribe service operation with an Immediate Report Indication that triggers the UDM to immediately return the subscribed data if the corresponding feature is supported by both the AMF and the UDM.
3. If the "SMS supported" indication is included in the Registration Request, the AMF checks in the SMS Subscription data that was received in step 2 whether the SMS service is allowed to the UE. If SMS service is allowed and the UE context received in step 2 includes an available SMSF of the serving PLMN, the AMF activates this SMSF Address and continues the registration procedure. If SMS service is allowed but an SMSF of the serving PLMN was not received in step 2, the AMF discovers and selects an SMSF to serve the UE as described in clause 6.3.10 of TS 23.501 [2]. If there are multiple SMSF instances in the network and the SMS delivery to Emergency Response Centre is only supported by certain SMSF instances, the AMF selects the SMSF instances that supports the SMS delivery to Emergency Response Centre (e.g. based on the SMSF profile that indicates the support of SMS delivery to Emergency Response Centre).
4. Step 15 to step 20 of the Registration procedure in Figure 4.2.2.2.2-1 are performed.
5. The AMF invokes Nsmsf_SMService_Activate service operation from the SMSF. The invocation includes AMF address, Access Type, RAT Type, Trace Requirements, GPSI (if available) and SUPI. AMF uses the SMSF Information derived from step 3. Trace Requirements is provided if it has been received by AMF as part of subscription data. When the AMF determines the UE has the MPS for Messaging indication in UE context as specified in step 2, the AMF includes a Message Priority header with a value appropriate for MPS to indicate priority treatment.
6. The SMSF discovers a UDM as described in clause 6.3.8 of TS 23.501 [2].
7a. If the UE context for the current Access Type already exists in the SMSF, the SMSF shall replace the old AMF address with the new AMF address.
Otherwise, the SMSF considers this a Registration request from a new Access Type and the SMSF registers with the UDM using Nudm_UECM_Registration with Access Type. As a result, the UDM stores the following information: SUPI, SMSF identity, SMSF address, Access Type(s) in UE Context in SMSF data. The UDM may further store SMSF Information in UDR by Nudr_DM_Update (SUPI, Subscription Data, UE Context in SMSF data).
If the Nsmsf_SMService_Activate request contains two Access Types and one of them is already registered in the SMSF, the SMSF shall replace the old AMF address with the new AMF address for that Access Type. The SMSF shall then register the other Access Type with the UDM using Nudm_UECM_Registration request.
7b-7c SMSF retrieves SMS Management Subscription data (e.g. SMS teleservice, SMS barring list) using Nudm_SDM_Get and this requires that UDM may get this information from UDR by Nudr_DM_Query (SUPI, Subscription Data, SMS Management Subscription data). After a successful response is received, the SMSF subscribes to be notified using Nudm_SDM_Subscribe when the SMS Management Subscription data is modified and UDM may subscribe to notifications from UDR by Nudr_DM_Subscribe.
SMSF also creates a UE context to store the SMS subscription information and the AMF address that is serving this UE.
NOTE 2: The SMSF can, instead of the Nudm_SDM_Get service operation, use the Nudm_SDM_Subscribe service operation with an Immediate Report Indication that triggers the UDM to immediately return the subscribed data if the corresponding feature is supported by both the SMSF and the UDM.
8. The SMSF responds back to the AMF with Nsmsf_SMService_Activate service operation response message. The AMF stores the SMSF Information received as part of the UE context.
9. The AMF includes the "SMS allowed", indication and "SMS over NAS delivery to Emergency Response Centre supported/allowed" based on local policy (e.g. the function is supported in serving AMF/SMSF) to the UE in the Registration Accept message of step 21 of Figure 4.2.2.2.2-1 only after step 8 in which the AMF has received a positive indication from the selected SMSF.
The "SMS allowed" indication in the Registration Accept message indicates to the UE whether the network allows the SMS message delivery over NAS. The "SMS over NAS delivery to Emergency Response Centre allowed/supported" indicates to UE whether the network allows the SMS over NAS delivery to Emergency Response Centre. The AMF may also include the emergency number list to UE if the "SMS over NAS delivery to Emergency Response Centre allowed" is set.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.2.2.1.2 Attach procedure in EPS
| The SMS over NAS delivery to Emergency Response Centre is similar to 5GS procedure with the change that the MME takes to role of AMF/SMSF.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.2.2.2 IMS Registration procedure
| The following procedures are based on TS 23.204 [8] with the following additions (in bold underline). This procedure is considered when UE is in non-roaming.
Figure 6.2.2.2-1: Based on TS 23.204 [8], Figure 6.1: Registration procedure
1) The UE establishes IP connection.
2) At any time after the establishment of the IP connection, the UE registers at the S-CSCF according to the IMS registration procedures. Based on local policy (e.g. the function is supported in IMS/IP-SM-GW), the S-CSCF also indicates the support of SMS over IP delivery to Emergency Response Centre by including e.g. "+g.3gpp.smsip-delivery to Emergency Response Centre" parameter in the Feature-Caps field header in 200 (OK) response. The "+g.3gpp.smsip-delivery to Emergency Response Centre" parameter indicates to UE whether the network allows the SMS over IP delivery to Emergency Response Centre.
NOTE 1: For simplicity, not all messages between UE and S-CSCF and between S-CSCF and HSS are shown in detail.
3) S-CSCF checks the initial filter criteria retrieved from the HSS during the IMS registration procedure.
4) After successful IMS registration and based on the retrieved initial filter criteria, the S-CSCF informs the IP-SM-GW (AS) about the registration of the user. IMSI is informed to the IP‑SM‑GW (AS) when there is no MSISDN in the UE's IMS subscription profile.
5) The IP-SM-GW (AS) returns OK to the S-CSCF.
6) The IP-SM-GW (AS) sends IP-SM-GW Register Req to the HSS.
7) The HSS stores the received IP-SM-GW address if necessary or for MT-SMS without MSISDN (see clause 6.0a.2), uses it as an indication that the UE is available to be accessed via the IMS to trigger an Alert service centre message if the message waiting flag is set and responses to the IP-SM-GW (AS) with IP‑SM‑GW Register Res. IP‑SM‑GW gets the IMPU (SIP URI) for SMS delivery without TEL-URI from registration event package.
NOTE 2: IP-SM-GW Register Res can include the SC address to be used for this user in the subscriber data (see also clause 6.7).
NOTE 3: If the IP‑SM‑GW address stored in the HSS via registration procedure is not the same as the preconfigured IP‑SM‑GW address (if any), then the short message delivery attempted during registration can be unnecessarily delayed.
8) After successful registration of the IP‑SM‑GW address at the HSS the HSS checks whether message waiting data are stored and alerts all SCs using procedures described in TS 23.040 [13] (see also clause 6.5b).
Editor's note: Roaming scenario is FFS.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.7.2.1.1 Emergency Registration Procedure in 5GS
| Figure 6.7.2.1.1-1 shows Emergency Registration procedure in 5GS.
Figure 6.7.2.1.1-1: Emergency Registration Procedure in 5GS
The following are the required enhancement for Emergency Registration Procedure:
1: The UE sends a Registration Request to the AMF. The message includes:
- 5GS registration type set to "emergency registration".
- A new Authentication indication IE or 5GS update type IE with the SMS requested bit set to "SMS over NAS supported" and a new SMS authentication required bit set to true (acting as an authentication trigger).
1a. The AMF recognizes the Emergency Registration and the explicit authentication need for SMS over NAS support. The AMF overrides the default deferral policy and initiates the primary authentication procedure immediately with the AUSF/UDM. The AMF performs authentication procedures to interact with UDM to ensure that the UE's credentials are validated before enabling SMS services.
2: AMF uses the Nudm_UECM_Registration operation/Nudm_UEAuthentication/ Service/ Nudm_SDM_Get/Nudm_SDM_Subscription service operations to interact with UDM.
3: Upon successful authentication, the AMF derives the NAS security context and continues the registration procedure. The AMF provides the VPLMN Emergency Number List and 5GS registration result indicating "SMS allowed" bit that confirms successful authentication and SMS for routing to local PSAPs services are authorized. If primary authentication fails, the AMF sends registration accept message with 5GS registration result indicating SMS not allowed to notify the UE that SMS over NAS is not possible for emergency registration due to lack of a security context or authorization, recommending fallback to another mechanism, e.g. normal registration procedure enhanced with SMS for LBO routing to PSAP as described in clause 6.7.2.1.2.
4: The AMF sends Registration Accept message to the UE indicating Emergency services supported and SMS service allowed/disallowed in the 5GS registration result.
5: The UE stores 5GS registration result for UE allowed/disallowed in UE context.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.7.2.1.2 Registration Procedure (VPLMN LBO Supported) in 5GS
| Figure 6.7.2.1.2-1 shows Registration procedure in 5GS.
Figure 6.7.2.1.2-1: Registration Procedure (VPLMN LBO Supported) in 5GS
The following are the procedure for Normal Registration (Negotiating Conditional LBO Capability):
1: The UE sends a Registration Request to the AMF, including the UE network capability IE with the SMS for LBO Support indication.
2: AMF includes SMS for LBO Support indication when using the Nudm_UECM_Registration operation/ Nudm_SDM_Get/Nudm_SDM_Subscription service operations to interact with UDM.
3- 7: Same as step 3-Step 8 in Figure 4.13.3.1-1: Registration procedure supporting SMS over NAS in TS 23.502 [3].
8: The V-AMF determines network support of SMS over NAS and SMS for LBO routing.
9: If conditional LBO for PSAP-destined SMS is allowed, the AMF sends a Registration Accept message to the UE, including the SMS for LBO Allowed indicator and the VPLMN Emergency Number List. The AMF may optionally include the visited SMSC Address IE.
9a: If conditional LBO for PSAP-destined SMS is not allowed, the AMF sends a Registration Accept message to the UE, omitting the MO PSAP SMS LBO Allowed indicator. The AMF includes the VPLMN Emergency Number List and may optionally include the visited SMSC Address IE.
10: The UE stores SMS allowed indication and SMS for LBO allowed indication (if available) in UE context.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.7.2.1.3 MO SMS over NAS Delivery in 5GS (VPLMN LBO Supported)
| Figure 6.7.2.1.3-1 shows MO SMS over NAS Delivery using LBO routing to local PSAP.
Figure 6.7.2.1.3-1: MO SMS over NAS delivery using LBO routing to local PSAP
The following are the procedure for MO SMS Delivery (Conditional LBO Trigger):
1: The UE performs Registration procedure as in Figure 6.7.2.1.2-1, or Emergency registration procedure in Figure 6.7.2.1.1-1.
2: The UE generates the SMS-SUBMIT TPDU with the local emergency number in the TP-DA field.
3: Based on SMS for LBO Allowed indication received and stored in UE context in step 1 and detected local emergency number, the UE sends SMS over NAS using an Uplink NAS Transport message with emergency indication or a new Payload container type IE, e.g. set to SMS for LBO routing.
4: The AMF receives the message and detects the emergency indication. The AMF overrides the default policy and selects a locally configured V-SMSF. If the AMF receives the "emergency indication" for MO SMS routing to local PSAP but the UE was not explicitly granted the SMS-LBO Allowed indication during registration (step 9 in Figure 6.6.2.1.2-1), the AMF shall ignore the emergency indication and follow the default Home Routed policy as described in Figure 6.7.2.1.4-1.
5: The AMF sends Nsmsf_SMService_UplinkSMS message including SMS-TPDU payload for LBO routing and SMS for emergency indication to the V-SMSF.
6: The V-SMSF determines that the MO SMS is for LBO routing.
7: The V-SMSF forwards MO SMS to the V-SMSC.
8: The V-SMSC identifies the MO SMS for LBO and routes it to local PSAP based on emergency number, emergency service type and location information, e.g. AML, indicated in the SMS-TPDU payload.
|
2b4a92fb259f5b8c209f1c52ea2cc338 | 23.700-65 | 6.7.2.1.4 MO SMS over NAS (VPLMN LBO Not Supported/Allowed - Home Routed Fallback)
| If MO SMS for LBO routing to local PSAP is not supported, the visited AMF applies the Home Routed policy. Figure 6.7.2.1.4-1 shows MO SMS over NAS Delivery in 5GS UE (VPLMN LBO Not Supported/Allowed- Home Routed Fallback).
Figure 6.7.2.1.4-1: MO SMS over NAS Delivery in 5GS (VPLMN LBO Not Supported/Allowed - Home Routed Fallback)
The following are the procedure for MO SMS over NAS Delivery in 5GS (VPLMN LBO Not Supported/Allowed - Home Routed Fallback) in Figure 6.7.2.1.4-1:
1: The UE performs Registration procedure as in Figure 6.7.2.1.2-1, or Emergency registration procedure in Figure 6.7.2.1.1-1.
2: The UE generates the SMS-SUBMIT TPDU with the local emergency number in the TP-DA field. If the UE received the visited SMSC Address IE in Registration procedure, the UE includes this address in a new dedicated routing field within the SMS TPDU.
3: Without SMS for LBO Allowed indication stored in UE context, the UE sends SMS over NAS using an Uplink NAS Transport message to the AMF.
4: The AMF receives the message and follows the default Home routed roaming policy, selecting the Home SMSF.
5: The visited AMF transmits/handles the SMS-RPDU which contains the SMS-TPDU and forwards the received SMS-RPDU message to the visited SMSF by invoking the Nsmsf_SMService_UplinkSMS service operation, including the full SMS message.
6: The V-SMSF retrieves the SMS-TPDU payload from SMS-RPDU and forwards SMS-TPDU payload to H-SMSF. The H-SMSF determines that the MO SMS TPDU is with emergency number and forwards MO SMS to the H-SMSC.
7: The H-SMSC identifies the MO SMS for routing to local PSAP in VPLMN based on emergency number, emergency service type and location information, e.g. AML, indicated in the SMS-TPDU payload. The H-SMSC also inspects the SMS TPDU payload for the visited-SMSC Address (if included by the UE). If present, the H-SMSC uses this address for direct routing to the visited SMSC Address for routing to local PSAP. If the visited SMSC Address is not present, the H-SMSC inspects the TP-DA and uses its global routing policy to route the message back to the appropriate visited SMSC for routing to local PSAP.
8: The H-SMSC routes the message to the V-SMSC.
9: The V-SMSC identifies the MO SMS and routes it to local PSAP based on emergency number, emergency service type and location information, e.g. AML, indicated in the SMS-TPDU payload.
Editor's note: It is FFS how to support Callback MT SMS over NAS delivery for local routing via V-SMSF.
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 1 Scope
| This TR provides guidelines for This TR provides guidelines for SEAL services usage for the benefit of the Application providers (developers/ application service provider) and application ecosystem partners (e.g., third party application platform providers).
It includes the overall introduction of SEAL layer from the aspects :1) SEAL role and responsibility within 3GPP exposure system, 2) the relationship between SEAL and with external SDO, 3) the introduction of SEAL services from the aspects including value, use case and advantages, and 4) deployment of SEAL entities and business relationship among stakeholders.
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 2 References
| The following documents contain provisions which, through reference in this text, constitute provisions of the present document.
- References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific.
- For a specific reference, subsequent revisions do not apply.
- For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.
[1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications".
[2] ……
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 3 Definitions of terms, symbols and abbreviations
| |
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 3.1 Terms
| For the purposes of the present document, the terms given in 3GPP TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in 3GPP TR 21.905 [1].
For the purposes of the present document, the terms given in clause 3 of 3GPP TS 23.222 [2] and clause 3 of 3GPP TS 29.222 [3] shall also apply.
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 3.2 Symbols
| For the purposes of the present document, the following symbols apply:
<symbol> <Explanation>
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 3.3 Abbreviations
| For the purposes of the present document, the abbreviations given in 3GPP TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in 3GPP TR 21.905 [1].
For the purposes of the present document, the abbreviations given in clause 3 of 3GPP TS 23.222 [2] and clause 3 of 3GPP TS 29.222 [3] shall also apply.
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 4 Overview of SEAL
| |
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 4.1.1 General description
| The services at this layer defined in 3GPP TS 23.434 [3], 3GPP TS23.435 [4], 3GPP TS23.436 [5], 3GPP TS23.437 [6], 3GPP TS23.438 [7], 3GPP TS23.482 [12].
As figure 4.1.1-1 shows, SEAL is on the 3GPP services layer Services and provides the value-added services to enable application layer to utilize 3GPP network and UE capabilities in an easy-to-use way. SEAL services are generic to any application type.
Services at this layer are created by the combination use of different network connection layer capabilities from one or multiple Network Functions deployed in telecom system, may potentially combining the resources/information of SEAL client on the UE.
Services exposed at this layer normally is larger granularity than the services exposed by single network function at network connection layer and provide an application layer meaningful function. It could simply the application layer implementation compared to directly consume network connection layer services by invoking APIs one by one.
SEAL layer includes SEAL services server(s) and SEAL client(s).
- SEAL services server(s): it is implemented in network by MNO or 3rd part and are authorized to (1) provide services to application and SEAL client, and (2) consume 3GPP network services by invoking network APIs.
- SEAL client(s): it is implemented in UE and can be provided by same organization as SEAL service server or other origination e.g., UE vendor/OS and are authorized to (1) provide services to consumers by APIs, (2) consuming network services by invoking APIs from SEAL service server (3) trigger UE modem to initiate NAS procedures to request network capability so as to provide service result to application.
•
4.1.1-1: Layered representation of SEAL service layer in 3GPP network exposure system
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 4.1.2 The interaction with rest part of 3GPP system
| Figure 4.1.2-1 Service based Architecture of 3GPP system supporting SEAL services
As the figure 4.1.2-1, SEAL services which are part of the 3GPP system are deployed either by the MNO or third party.
SEAL server(s) invokes service APIs exposed by 3GPP core network functions directly over the 3GPP core network SBA when deployed as trusted AF in MNO's trusted domain as figure 4.x.1-1.
NOTE 1: In the real case, SEAL server(s) may invoke service APIs exposed by 3GPP core network functions via NEF as a non-trusted AF e.g., when deployed by the third party.
NOTE2: SEAL server(s) when deployed as trusted AF in MNO's trusted domain may also invoke APIs from 3GPP core network functions via NEF.
SEAL server(s) invokes service APIs exposed by the OAM system over the OAM SBMA when deployed by MNO as specified in 3GPP TS28.533 [14].
NOTE 3: In the real case, SEAL server(s) may invoke service APIs exposed by 3GPP core network functions via AEF (e.g., when deployed by the third party, or even by MNO).
SEAL clients take two different responsibilities:
1) consumer of SEAL server API(s)
2) providing required terminal side information/resource to support the SEAL server function(s)
SEAL services APIs are provided and exposed by MNOs or third party, and they are consumed by the Application specific layer (VAL layer) or vertical application enabler layer (e.g., VAE server, UAE server), Edge enabler layer (e.g., EES, ECS), SEAL client(s) or third party's platform.
The interactions between the vertical application server(s) and SEAL server(s) for discovery, authentication and authorization are supported by CAPIF as specified in 3GPP TS 23.222 [9]. The VAL server acts as CAPIF's API invoker and SEAL server acts as CAPIF's API exposing function.
NOTE 4: It is up to application developer's implementation how to invoke the SEAL server's APIs in their application layer software.
NOTE 5: The API invokers as defined in 3GPP TS 23.222 [9] can either be on the network side (Application server) or on the UE side (e.g., SEAL clients, applications on UE), to use SEAL server API(s).
4.x SEAL Entities
Editor's Note: This clause will describe the service entities (SEAL servers and SEAL Clients) in SEAL layer.
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 5 The usecases and advantages to applications
| |
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 6 Deployment of SEAL entities and business relationship among stakeholders
| Editor's Note: Deployment alternatives by stakeholders and corresponding business relationship
Annex A:
Relationship between SEAL Enablers and external organizations
Editor's Note: This clause will describe how SEAL layer could be consumed in different vertical scenarios based on the analysis in solution#6 in the 3GPP TR 23.700-35.
Annex B:
Change history
Change history
Date
Meeting
TDoc
CR
Rev
Cat
Subject/Comment
New version
2025-10
SA6#69
S6-254792
TS skeleton
0.0.0
2025-10
SA6#69
Implementation of the following pCRs approved by SA6: S6-254660, S6-254661, S6-254662, S6-254692
0.1.0
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 4.1 The role and responsibility of SEAL within 3GPP system
| Editor's Note: This clause will describe the general role and responsibilities of SEAL layer
|
835f30c35d1989224fb1e21b58c42c99 | 23.949 | 5.1 Summary of SEAL services
| Editor's Note: Brief summary of the value of each the SEAL services.
This clause is to address the overall summary of the values and beneficial use case(s) of each SEAL service.
The supporting of the following are the key services that make SEAL layer distinct with other non-3GPP basis developer API Platform:
- NSCE/NRM/SEAL-DD which are specific to the usage of 5G/4G network capabilities.
- E-MMTel enabler which is specific the usage of IMS system capabilities.
- LMS utilize both 3GPP based position and non-3GPP based position.
- AI/ML enabler could support AI/ML service related to mobile UE, leveraging on the 3GPP network-for-AI capabilities
Service type
Service description
Benefits to application provides
NRM
• Provide value added APIs by abstrating the capabilties served by core network layer entities MB-SMF/PCF/NEF APIs/SCEF/BMSC, to enable third party to use 3GPP QoS service in a simple way
Simply the usage of App providers (APP deveolper, APP owner) of 3GPP network capabilities to:
- Improve qualitiy of application layer communication by using NRM API to tigger the 3GPP network connection with QoS supporting, compared to application layer communication over best effort 3GPP network connection.
SEAL DD
• Provide Caching and Distribution between application content server and mobile UE by using various 3GPP network capabilities designed for different type of application data to optimization purpose. For example reliable transmission mechanism specific to URLLC data.
Futher improve QoE of the applications when using Caching and Distribution service, by supporting the usage of 3GPP network capabilities(including control plane and user plane) for QoS optimization
NSCE
• Provide the API services to enable third party to retrive the slice SLA requirement automaticly, order slice product from MNO system, mornitor/change slice SLA, by aggregating the API services from OAM and Core network, SEAL layer services (service KQI information)
Provide the required functions(e.g deriving and recomending slice SLA to let a slice customer e.g., enterprise knows how much resource is required for there services) to enable third party to perform self-management operation on their dedicated slice, when NaaS is provided by operator OAM.
AI/ML enabler
• AI/ML AIMLE provides APIs for enabling and managing AI/ML services
• It leverages the network capabilities of 5G systems to support secure and efficient connection for mobile application to utilize AI services from MNO,
It could enable an MNO operator to use their AI capabilities to launch the new business to AI applications.
A mobile AI application can gain from offload the AI task to another node to promote the model performance or latency performance.
LMS
LMS provides location management related functionality and location based services to the 3rd party applications. It utilizes the core network related location capabilities and value add on top of it in-order to provide a enriched location services.
Availability of enriched Location information from core network along with the GPS co-ordinates. provides readily available supplementary location based services such as geofencing, location deviation monitoring etc., apart from reporting and fetching of location information.
Group management
Group management service provides the group management related operations considering the needs of multiple vertical applications. Apart from regular CRUD operations it provides the value add services like location based group creation, group regrouping, group member registration etc.
Re-usability : Single group management framework catering to multiple vertical needs easing the development time.
Configuration management
Configuration management is to enable ASP to provide configuring data applicable to different vertical applications.
Re-usability : Single configuration management framework catering to multiple vertical needs easing the development time.
Digital asset
Provides application enablement to Metaverse application by supporting following functionalities:
• Digital asset profile management: which includes creating, updating, retrieving and deleting the digital asset profile
• Digital asset discovery
Digital asset media management – which includes uploading and downloading the digital asset media.
Capabilities to re-use the digital asset across multiple Metaverse applications (e.g. digital assets server provided by the MNO, which is trusted by different Metaverse applications to manage digital assets).
Spatial Anhor enabler and spitial mapping enabler
Offers the platform for managing the spatial anchors and spatial maps for metaverse services. Enables the re-use of spatial anchors and spatial maps across multiple verticals and thus improving the metaverse experience for the end users.
Capability to re-use the spatial anchors and maps across multiple verticals.
Provides the insight on the spatial anchor usage to improve the effeciency
5.y Usecases and advantages for SEAL Service X
Editor's Note: Details about the value of each SEAL services and how it brings advantages to applications
|
97c399698bf3bda50453286bc4472172 | 28.881 | 1 Scope
| The present document …
|
97c399698bf3bda50453286bc4472172 | 28.881 | 2 References
| The following documents contain provisions which, through reference in this text, constitute provisions of the present document.
- References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific.
- For a specific reference, subsequent revisions do not apply.
- For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.
[1] 3GPP TS 28.312: "Management and orchestration; Intent driven management services for mobile networks".
[2] 3GPP TS 28.541: "Management and orchestration; 5G Network Resource Model (NRM); Stage 2 and stage 3".
[3] 3GPP TS 29.572: "5G System; Location Management Services; Stage 3"
[4] 3GPP TS 38.304: "NR; User Equipment (UE) procedures in Idle mode and in RRC Inactive state".
[5] 3GPP TS 38.331: "NR; Radio Resource Control (RRC); Protocol specification".
[6] 3GPP TS 28.537: “Management and orchestration; Management capabilities”.
[7] 3GPP TS 38.300: “NR; NR and NG-RAN Overall description; Stage-2”.
[8] 3GPP TS 28.554: "Management and orchestration; 5G end to end Key Performance Indicators (KPI)".
[9] 3GPP TR 28.914: "Study on intent driven management service for mobile network phase 3".
[10] 3GPP TS 28.105: "Management and orchestration; Artificial Intelligence/ Machine Learning (AI/ML) management".
[11] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications".
|
97c399698bf3bda50453286bc4472172 | 28.881 | 3 Definitions of terms, symbols and abbreviations
| |
97c399698bf3bda50453286bc4472172 | 28.881 | 3.1 Terms
| For the purposes of the present document, the terms given in TR 21.905 [11] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in TR 21.905 [11].
example: text used to clarify abstract rules by applying them literally.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 3.2 Symbols
| For the purposes of the present document, the following symbols apply:
<symbol> <Explanation>
|
97c399698bf3bda50453286bc4472172 | 28.881 | 3.3 Abbreviations
| For the purposes of the present document, the abbreviations given in TR 21.905 [11] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in TR 21.905 [11].
<ABBREVIATION> <Expansion>
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4 Use cases and potential solutions for new areas
| 4.1 Use case #1: Enhancement of radio service delivering and assurance scenarios
4.1.1 Description
In 3GPP TS 28.312 [1], the existing use case and requirements for intent containing an expectation for delivering a radio service is described in clause 5.1.2. The RadioServiceExpectation is defined to represent MnS consumer's expectations for radio service delivering and assurance in the specified area. However, following scenarios are not supported:
- MnS consumer expresses the radio service delivering and assurance expectation with service reliability information.
- MnS consumer expresses the radio service delivering and assurance intent expectation for a specified area described in the form of civic address. For example, MnS consumer wants to ensure the radio service targets (dLThptPerUETarget and dLLatencyTarget) for a specific civic address (e.g., the CivicAddress defined in clause 6.1.6.2.14 in TS 29.572 [3]).
- MnS consumer expresses the radio service delivering and assurance expectation for a specified assurance time duration. Using the concert as an example, MnS consumer expresses the intent indicating a radio service to be delivered and assured for two hours, or before concert, MnS consumer request to explore the supported maximum number of UE for the Radio Service which to be assured for two hours after it being delivered.
4.1.2 Potential requirements
REQ-IDMS_RadioServiceIntent -CON-1: The intent driven MnS producer for radio service should have capabilities enabling the MnS consumer to express service reliability requirements.
REQ-IDMS_RadioServiceIntent -CON-2: The intent driven MnS producer for radio service should have capabilities enabling the MnS consumer to express radio service delivering and assurance for a specified area described in the form of civic address.
REQ-IDMS_RadioServiceIntent -CON-3: The intent driven MnS producer for radio service should have capabilities enabling the MnS consumer to express radio service assurance time duration requirements.
4.1.3 Potential solutions
This solution proposes to reuse and enhance the existing RadioServiceExpectation defined in 3GPP TS 28.312 [1].
Enhancement Aspect1: Add following attributes as the ExpectationTargets for the RadioServiceExpectation to enable the MnS consumer to express service reliability requirements
- ReliabilityTarget, it represents the reliability target for the radio service that the intent expectation is applied.
Enhancement Aspect2: Add following attributes as the ObjectContexts for the RadioServiceExpectation to enable the MnS consumer to express radio service delivering and assurance requirements for a specified area described in the form of civic address.
- CivicAddressContext, the coverage areas for the Radio Service that the intent expectation is applied in the form of civic address (e.g. streets, districts, seats, etc.). The detailed definition for civic address reuses the CivicAddress Data Type defined in clause 6.1.6.2.14 in TS 29.572 [3].
Enhancement Aspect3: Add following attributes as the Expectation Contexts for the RadioServiceExpectation to support assurance duration information:
- AssuranceDurationContext, it describes the time duration at which the radio service instance should be scheduled to be delivered and available. The type is integer and unit is hour.
4.1.4 Evaluation of potential solutions
TBD
4.2 Use case #2: Enhancement of radio network performance assurance scenarios
4.2.1 Description
In 3GPP TS 28.312 [1], the existing use case and requirements for intent containing an expectation on radio network performance to be assured and intent containing an expectation for RAN energy saving are described in clause 5.1.5 and clause 5.1.7. The RadioNetworkExpectation is defined in clause 6.2.2.1.1 to represent MnS consumer's expectations for radio network delivering and performance assurance. However, following scenarios are not supported:
- MnS consumer expresses radio network performance assurance expectation for a specific RAN feature (e.g., RedCap). For example, MnS consumer may express the radio network performance targets (e.g. weakRSRPRatioTarget, highUlPrbLoadRatioTarget and aveULRANUEThptTarget) to be assured for RedCap UEs in the specified areas. The detailed RAN feature need further investigation on TS 38.304 [4] and TS 38.331[5].
- MnS consumer expresses relative values for several performance targets for the specified areas. For example, MnS consumer may express the expectation on RAN energy consumption reduction ratio (as percentage) for RAN SubNetwork that the intent expectation is applied to illustrates the difference between the energy consumption before and after performing energy saving actions.
4.2.2 Potential requirements
REQ-IDMS_RadioNetworkIntent -CON-1: The intent driven MnS producer for radio network should have capabilities enabling the MnS consumer to express radio network performance assurance for a specific RAN feature
4.2.3 Potential solution
This solution proposes to reuse and enhance the existing RadioNetworkExpectation defined in 3GPP TS 28.312 [1] to enable the MnS consumer to express radio network performance assurance for a specific RAN feature.
Enhancement Aspect1: Add following attributes as the ObjectContexts for the RadioNetworkExpectation:
- rANFeatureContext, it represents the expected specific RAN feature for RAN Subnetwork that the intent expectation is applied. Following are the allowed values for RAN features:
- REDCAP: it represents the support of Reduced Capability (RedCap) and enhanced Reduced Capability (eRedCap) NR devices as defined in TS 38.300 [7].
- XR: it represents the support of eXtended Reality (XR) services that require high data rate and low latency communications as defined in TS 38.300 [7].
- AERIAL_UE_COMMUNICATION: it represents the Aerial UE Communication as defined in TS 38.300 [7].
- V2X: it represents the V2X communication as defined in TS 38.300 [7].
Note: allowed values can be exteneded based on RAN specifications.4.2.4 Evaluation of potential solutions
TBD
4.2.4 Evaluation of potential solutions
TBD
4.3 Use case #3: Assisting and reporting intent decomposition across intent handling functions
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.3.1 Description
| An intent-driven MnS producer can translate an intent and decompose it into one or more new intents. If these newly generated intent(s) cannot be handled by the intent-driven MnS producer, it can submit them to another intent-driven MnS producer(s). In this case, the former intent-driven MnS producer acts as the consumer of the latter intent driven MnS producer(s), to which the newly generated intents are submitted.
The issue arises when the decomposition of an intent into new intents is carried out by an intent driven MnS producer, these decomposed intents may be submitted to intent-driven MnS producers of managed domains that are not desired by the original intent-driven MnS consumer. The intent-driven MnS consumer needs to be able to assist the intent driven MnS producer regarding the decomposition of intents to other MnS producers. Furthermore, the intent driven MnS consumer needs to receive a report regarding this decomposition. This is also important for troubleshooting an intent-driven MnS producer, to check whether it is acting as expected.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.3.2 Potential requirements
| REQ-Intent_DCMP-1: The intent driven MnS producer should have the capability to allow an MnS consumer to specify a context to be able to assist the intent-driven MnS producer regarding the decomposition of an intent to other MnS producers.
REQ-Intent_DCMP-2: The intent driven MnS producer should have the capability to allow an intent-driven MnS consumer to receive a report regarding the decomposition of an intent to other MnS producers.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.3.3 Potential solutions
| |
97c399698bf3bda50453286bc4472172 | 28.881 | 4.3.3.1 Potential solution #1
| This solution proposes to reuse and enhance the existing intent expectations and intent report information models defined in 3GPP TS 28.312 [1].
Intent expectations are enhanced with:
• A new expectation context that specifies the constraints applicable in the intent decomposition performed by the intent-diven MnS producer, such as a list of Intent Handling Functions identity among those under MnS consumer’s authorization to which the intent decomposition is not recommended by the intent-driven MnS consumer.
The IntentReport IOC in 3GPP TS 28.312 [1] is enhanced with following information:
• The identity of Intent Handling Functions to which the decomposed intents are sent.
• The identity of each intent and expectation resulting from the decomposition.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.3.4 Evaluation of potential solutions
| TBD
4.4 Use case #4: Intent traceability
4.4.1 Description
An intent consumer (owner) submits an intent to a single intent handling function. In some cases, in order to fulfil the intent an intent handling function may need to submit additional intent(s) to other intent handling functions. Such handling can occur multiple times across intent handling functions across multiple management and/or domain layers.
An operator/administrator needs visibility of the relationships between intents which have been created by the system. Since these are being created within/by the management systems (and not explicit consumers the operator may implement) it’s important to know where they came from to allow ‘trace-back’ to the original consumer intent which started the cascade of subsequent intents. There is however no identified method, standardized or otherwise, which allows for such traceability.
Since intents can result in new intent(s) to multiple intent handling functions, likely with different implementations, it is insufficient to rely on external mechanisms such as logging or local network management audit tools to trace the intent. The information identifying each intent handling function which has handled the intent must be preserved along with the intent itself and accessible/meaningful within the content of each intent handling function. The intent consumer (owner) (e.g. Consumer A in figure 4.4.1-1) should be enabled to indicate whether they agree that their information be propagated beyond the recipient intent producer (e.g. Producer 1 in figure 4.1.1-1) to other intent producers (e.g. Producer 2 or 3 in figure 4.4.1-1).
The following figure provides an overview of such information and its handling:
Figure 4.4.1-1: Intent traceability information handling
4.4.2 Potential requirements
REQ-Intent_Trace-1: The intent driven MnS producer should provide information in the intent to identify that an intent has been handled by a particular intent handling function.
REQ-Intent_Trace-2: The intent driven MnS producer should provide information (as defined in REQ-Intent_Trace-1) to identify any subsequent intents created by it as part of fulfilment.
REQ-Intent_Trace-3: The intent driven MnS producer should provide information (as defined in REQ-Intent_Trace-1) identifying the intent handling functions to subsequent intent(s) to allow traceability of the intent across multiple intent handling functions.
REQ-Intent_Trace-4: The intent driven MnS producer should provide a capability allowing the intent MnS consumer (owner) to indicate whether the MnS consumer agrees that their information can be propagated in case of intent decomposition beyond the recipient intent producer to other intent producers.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.4.3 Potential solutions
| 4.4.3.1 Potential solution #1
To address REQ-Intent_Trace-1, REQ-Intent_Trace-2 and REQ-Intent_Trace-3 this solution proposes adding a new data structure to the intent model definition to define the information to trace the decomposition of an intent into subsequent intents. The information is updated and propagated as new intent(s) are created as part of fulfilment.
The content proposed for the intent model definition would include
• the identity of each intent handling function which decomposed the intent
• the identity of each intent resulting from the decomposition
The handling of the information would be:
- a new intent created by an MnS Consumer is not required to populate this value, i.e. default is empty string.
- upon receipt of an intent, the Intent Handling Function updates the information to indicate it handled the intent by adding its identity to the incoming intent
- in the event the Intent Handling Function determines intent decomposition is required it create new intent(s) and propagates the information from the source intent to the new intent(s)
The visibility of this information is subject to the same access control as any other information exchanged between an authorised MnS consumer and producer. For example, the originating Intent Owner Consumer A in Figure 4.4.1-1 will have access/visibility to its intents, including Intent A at Producer 1. Producer 1 is an authorised MnS consumer of and will have access/visibility of its intents at Producer 2 and Producer 3, including Intent A.1 and A.2 resulting from the decomposition of Intent A. When no further decomposition is required, no further update is made to the information.
4.4.3.1 Potential solution #2
To address REQ-Intent_Trace-4, this solution proposes adding a new attribute to the intent model definition to allow the MnS consumer indicate its preference for the propagation of intent traceability information.
The content proposed for the intent model definition would include:
• an attribute to indicate intent decomposition, e.g. ‘includeTraceInfo’ of type boolean
• default value of True
• value would be invariant
• configurable by MnS Consumer
The handling of includeTraceInfo would be:
- a new intent will have includeTraceInfo default value of True
- MnS Consumer may modify value of includeTraceInfo
- if includeTraceInfo =True, the recipient MnS Producer may include intent traceability information from the originating intent.
- if includeTraceInfo =False, the recipient MnS Producer may not include intent traceability information from the originating intent
- In the event of intent decomposition, the value of includeTraceInfo may or may not be propagated to subsequent intents at the discretion of each MnS Producer, i.e. in its role as MnS Consumer towards the subsequent MnS Producer(s).
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.4.4 Evaluation of potential solutions
| TBD
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.5 Use case #5: Invariant Guidance in Intent Contexts
| 4.5.1 Description
TS 28.312 describes a scenario where for an instantiated intent, the intent handler may need to decompose the intent into multiple derivative intents, each to be fulfilled by a separate other intent handler. There may be contexts in the intent which the MnS consumer desires that the other intent handlers understand them exactly as they were provided, i.e., without any modification by the first intent handler in the decomposition process.
As an example, the MNO may, for some decomposition use-cases, want to only use certain certified hardware (e.g., with certain security, privacy, energy consumption, other types of quality guarantees) to be used in the fulfillment of RAN or CN operations. This additional guidance to use this specified hardware or the quality constraints on the resources may be provided as part of the intent context.
The MnS consumer should be enabled to indicate the context which should be transmitted to other intent handlers without modification.
4.5.2 Potential requirements
REQ-Intent_InvarGui-1: The intent driven MnS should include a capability enabling the MnS consumer to indicate the requirements, goals and contexts which the MnS consumer desires to be copied into decomposed intents and transmitted to other intent handlers without modification.
4.5.3 Potential solutions
TBA
4.5.4 Evaluation of potential solutions
TBA
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.6 Use case #6: Intent Interpretation Assistance Information
| 4.6.1 Description
A similar intent can be instantiated on several intent handlers, e.g., on several management systems each responsible for a for a different domain. TS28.312 supports intent negotiation that enables interaction between the MnS producer and MnS consumer to clarify the ability to fulfill intent requirements that may otherwise not be realizable owing to their abstract nature. For the same intents that are sent to several MnS producers, negotiation with the several MnS producers to clarify the intents can be resource wasteful.
The intent driven management service can minimize the number of negotiation exchanges while allowing effective interpretation of intents by reusing the interpretation information to other MnS producers. The MnS consumer compiles prior interactions with intent handlers, e.g. previous IntentNegotiationReports or intentNegotiationConsumerFeedback, and shares those prior interactions as intent interpretation assistance information with the other intent handlers.. The Mns consumer should be enabled to provide this intent interpretation assistance information that can be used by the intent handler to support the interpretation of the intent.
Note: The intent interpretation assistance information can be visualized as a “frequently asked questions (FAQ)” section used in many human-oriented service descriptions (e.g. on service providers’ websites).
4.6.2 Potential requirements
REQ-Intent_ Interpretation_Assistance_1: The intent driven MnS producer should have the capability enabling the MnS consumer to provide information containing prior intent negotiations to other intent handlers that can then be used by the intent handler to support the interpretation of the intent.
4.6.3 Potential solutions
TBA
4.6.4 Evaluation of potential solutions
TBA
4.7 Use case #7: Enhancement of intent exploration
4.7.1 Description
In 3GPP TS 28.312 [2], the existing use case and requirements for intent Exploration is defined in clause 5.3.5. The IntentExplorationReport is defined in clause 6.2.1.3.19 to represent the intent exploration result, which includes the list of expectation exploration results. However, following scenarios are not supported:
- MnS consumer requests intent handling function to periodically report the best target values. For example, MnS consumer request intent handling function to report the best value for UENumberTarget of RadioServiceExpectation by the period of an hour. In this case, MnS producer need to execute the exploration process and send the intent exploration report with the best values for UENumberTarget to MnS consumer every hour. As the intent exploration process is complex and time-consuming (e.g. MnS producer may perform simulation activities to explore the best values), the MnS consumer needs to obtain the status (e.g., RUNNING, FINISHED, FAILED) of the intent exploration process for specific intent expectation (s).
- MnS consumer requests intent handling function to explore best target values for individual objects (e.g. area, cells) indicated in one IntentExpectation. For example, MnS consumer request to obtain the best values for UENumberTarget applied to ExpectationObject which represents a list of cells. In the request, MnS consumer wants to obtain the best values for UENumberTarget for each cell in addition to the best value for the list of cells. In this case, MnS producer needs to provide the following information in the intent exploration report:
- The best UENumberTarget value for the list of cells represented by the ExpectationObject.
- The best UENumberTarget value for each item of the list of cells represented by the ExpectationObject.
4.7.2 Potential requirements
REQ-IDMS-IntentExploration-CON-1: The intent driven MnS producer should have the capability enabling the MnS consumer to request to periodically obtain the exploration report.
REQ- IDMS-IntentExploration-CON-2: The intent driven MnS producer should have the capability enabling the MnS consumer to obtain the best target values for individual object represented by the ExpectationObject in an IntentExpectation.
4.7.3 Potential solution
This solution proposes to reuse and enhance the existing Intent and IntentReport IOC defined in 3GPP TS 28.312 [1].
Enhancement Aspect1: Extend the IntentReportControl <<dataType>> with following aspects:
- Extend the attribute "observationPeriod" in IntentReportControl <<dataType>> to indicate the time period for which the IntentExplorationReport is observed and reported. If the attribute "observationPeriod" is presented in the Intent instance with intentMgmtPurpose = EXPLORATION, MnS producer periodically explore the best target values and send the IntentExplorationReport to MnS consumer at the end of each observation period. If the attribute "observationPeriod" is absent in the Intent instance with intentMgmtPurpose = EXPLORATION, MnS producer performs the exploration process one time and send the IntentExplorationReport after finishing exploration process.
Enhancement Aspect2: For the ExpectationTarget <<dataType>> used for the attribute targetExplorationResult, the targetContext should be used the targetContext can be cellContext, coverageAreaPolygonContext. In this case, multiple TargetExplorationResult instances can be created for the same target name (e.g. UENumberTarget) with different cellContexts or coverageAreaPolygonContext. Following is one example for TargetExplorationResult instances in one IntentExplorationResult.
[UENumberTarget, IS_LESS_THAN, 10]
[UENumberTarget, IS_LESS_THAN, 4, Cell1]
[UENumberTarget, IS_LESS_THAN, 3, Cell2]
[UENumberTarget, IS_LESS_THAN, 3, Cell3]
Enhancement Aspect3: extend the IntentExplorationReport <<dataType>> with following aspects:
- Add the attribute "expectationExplorationStatus" to represent the status of intent exploration. The allowed values can be NOT_STARTED, RUNNING, FINISHED, FAILED.
4.7.4 Evaluation of potential solutions
TBD
4.8 Use case #8: Support to express guarantee requirements in an intent
4.8.1 Description
In today’s highly dynamic and heterogeneous network environment, Intent-Driven Management Service (IDMS) is emerging as the foundational paradigm for advanced network operations and automation. Nevertheless, practical deployments reveal a critical capability gap: the absence of robust support for dynamic resource reservation and release.
Intent expression, translation, negotiation, and fulfilment are intrinsically coupled to resource allocation—e.g., bandwidth, capacity and spectrum. When the IDMS is used for communication service assurance in a specific time window, the MnS consumer should be allowed to express the guarantee requirements in the intent. Then, the MnS Producer can perform some actions (e.g., resource reservation) to guarantee the intent fulfilment in the future. Because these resources are highly time-variant, the current IDMS lacks the ability for the MnS Consumer to proactively reserve or release them on demand. This shortcoming results in inefficient resource utilisation and reduced flexibility in intent fulfilment. Additionally, after an MnS Producer has classified an intent as “FEASIBLE” following the initial feasibility check, subsequent changes in network resource can render the intent “INFEASIBLE” before intent fulfilment is requested without actions taken for guarantee. This temporal inconsistency, caused by the absence of continuous feasibility validation, will lead to fulfilment failures, undermining the reliability of intent-based operations and degrading overall network performance.
An intent defines one or more requirements, goals and constraints that should be ensured for the intent to be considered fulfilled. The requirements are evaluated against the state of the network during the intent fulfilment feasibility check. Even if the result is feasible, no guarantees are explicitly provided by the MnS producer to ensure that the requirements will be continuously fulfilled over time. While the MnS producer may find a solution that meets the requirements when the intent is accepted, it may later degrade due to over-allocation, faults, or other issues.
To tackle this issue, the concept of guaranteed requirements is introduced. Such requirements should be considered for forward-looking sustained compliance. The MnS Producer should be able to utilize different methods to guarantee the intent fulfilment in the future.
NOTE: The guaranteed requirements may apply to all or part of an intent - that is, to the entire intent or to specific expectations. This study will determine which aspects of an intent should be covered by guaranteed requirements.
Since in reality there is no way to guarantee something all the time (i.e., there is always a probability that a requirements will be breached), Guaranteed requirements are associated with a confidence level that expresses the probability of successfully in fulfilling the requirements. The confidence level is specified as a probability threshold and is provided by the MnS producer. How to express confidence may depend on the context of the intent, e.g. for a connectivity service-related intent, the confidence may be expressed as service availability and service reliability. How the MnS producer will provide the confidence level to the requirements is implementation specific.
Since guarantees may not extend indefinitely, the MnS consumer may specify a guarantee period over which the MnS producer provides the confidence level. The guarantee period begins when an intent is created or updated.
4.8.2 Potential requirements
REQ-IDMS_Guarantee-1: The intent driven MnS producer should have the capability to allow MnS consumer to express an intent to be guaranteed and which expectations requirement in an intent should be guaranteed.
REQ-IDMS_Guarantee-2: The intent driven MnS producer should have the capability to allow MnS consumer to specify the guarantee period.
REQ-IDMS_Guarantee-3: The intent driven MnS producer should have the capability to report to the MnS consumer about the confidence level of fulfilling the guaranteed expectations at time of reporting.
4.8.3 Potential solutions
To support to express guarantee requirements in an intent, the following enhancements are proposed:
Add a new attribute guaranteeIndicator in the Intent IOC to allow the MnS Consumer to specify Guarantee Requirements in the intent. guaranteeIndicator attribute can be on intent expectation level.
4.8.4 Evaluation of potential solutions
4.9 Use case #9: Intent handling capability configuration, registration and discovery
4.9.1 Description
4.9.1.1 Intent handling capability description
In TS 28.312 [1], the existing use case, requirements and solution (including IntentHandlingCapability <<dataType>>) for intent handling capability obtaining are defined, which allows the MnS consumer to query the intent handling capabilities for a specific intent handling function. The use case also describes that different intent handling functions are deployed to support different areas of the same intent expectation object domain. However, the IntentHandlingCapability <<dataType>> (including supportedExpectationObjectType and supportedExpectationTargetInfoList) is not allowed to describe the supported area information. In additional, for radio network intent, multiple intent handling functions can be deployed to support different radio access technologies (e.g. EUTRAN, NR) and/or different frequencies. It is important to describe the supported radio access technologies (e.g. and frequencies for the radio network intent handling functions.
Several optional intent negotiation functionalities (including Intent Feasibility check, Intent Exploration and Intent Fulfilment Negotiation) are introduced in the TS 28.312 [1]. One intent handling function may support all the negotiation functionalities or part of the negotiation functionalities. For example, intent handling function A support both Intent Feasibility check and Intent Exploration functionalities, while intent handling function B only support Intent Feasibility check functionality. So, it is important to allow MnS consumer to know which negotiation capabilities can be provided by a specific intent handling function.
4.9.1.2 Intent handling capability registration and discovery
In 3GPP TS 28.537 [6], the existing use case, requirements and solution (including MnSInfo IOC) for MnS discovery are defined, which allows MnS consumer to retrieve the list of MnS instances which can provide the intent driven management capability from MnS Registry. In this case, MnS consumer needs to send the query request to each intent handling function (represented by mnsAddress of the retrieved MnS instances from MnS Registry) to obtain the detailed intent handling capabilities.
However, in some scenarios, MnS consumer may need to request to retrieve a MnS instance with a specific intent handling capability from MnS Registry.
4.9.1.3 Intent handling capability configuration
As TS 28.312 [1] described, multiple intent handling functions maybe deployed to support different expectation objects or to support different areas of the same intent expectation object domain. Operator may need to configuration the intent handling capability for each intent handling function with different responsibilities. For example, intent handling function A is configured to support handling radio service intent in the Venue A, while intent handling function B is configured to support handling radio service intent in the Venue B.
4.9.2 Potential requirements
REQ-IDMS_IHCO-CON-1: The intent driven MnS producer should have capabilities enabling an MnS consumer to obtain intent handling capabilities of each intent handling function, including supported contexts.
NOTE: Examples of contexts to be included as part of the intent handling capabilities are the ObjectContexts and the ExpectationContexts as defined in clause 6.2.2.1 of 3GPP TS 28.312 [2].
REQ-IDMS_IHCO-CON-2: The intent driven MnS producer should have capabilities enabling an MnS consumer to obtain intent negotiation capability of a specific intent handling function.
REQ-IDMS_IHCO-CON-3: The 3GPP management system should have capabilities enabling the configuration of the intent handling capability for a specific intent handling function.
Editor’s Note: which content of the intent handling capability can be configured needs further investigation.
4.9.3 Potential solution
This solution proposes to reuse and enhance the existing IntentHandlingFunction defined in 3GPP TS 28.312 [2].
Enhancement Aspect1: Add following attributes for IntentHandlingCapability <<dataType>>:
- supportedObjectContextInfoList, it represents the list of specific ObjectContexts supported by the intent handling function. The new SupportedObjectContextInfo datatype includes the supportedObjectContextAttribute, supportedObjectContextCondition and supportedObjectContextValueRange attributes.
- supportedExpectationContextInfoList – it represents the list of specific ExpectationContexts supported by the intent handling function. The new SupportedExpectationContextInfo datatype includes the supportedExpectationContextAttribute, supportedExpectationContextCondition and supportedExpectationContextValueRange attributes.
Enhancement Aspect2: Add following attributes for IntentHandlingFunction <<IOC>>:
- supportedNegotiationFunctionalities, it represents the list of intent negotiation functionalities supported by the intent handling function. The type is ENUM with allowed values: FEASIBILITYCHECK, EXPLORATION and FULFILMENT_WITH_NEGOTIATION. If the value is absent, none of the intent negotiation capabilities are not supported by the intent handling function.
- intentHandlingScope, representing the scope of received intent that the intent handling function can support to address, the allowed values can be RAN, or CN.
4.9.4 Evaluation of potential solutions
TBD
4.10 Use Case#10: Radio network delivering in transient overload scenario
4.10.1 Description
This use case describes a scenario where a MnS consumer express intent containing an expectation for ensuring a radio network in transient overload scenarios (e.g., high-speed rail or subway systems) to a MnS producer. In transient overload scenarios, the high load caused by massive user access and bursty traffic typically lasts only a few seconds specifically when trains pass through a cell, while traffic volume remains extremely low during other periods.
In 3GPP TS 28.312 [1], the existing use case and requirements for intent containing an expectation for delivering a radio network is described in clause 5.1.2. The RadioServiceExpectation is defined to represent MnS consumer's expectations for radio network delivering in the specified area, and it serves to support service assurance efforts. However, transient overload scenario is not supported.
- MnS consumer expresses the radio network delivering expectation with transient load information. It can help MnS producer identify the actual high load information when these shot-duration, high-impact events occur thereby supporting network expansion decision-making.
4.10.2 Potential requirements
REQ-IDMS_RadioServiceIntent -CON-1: The intent driven MnS producer for radio network should have capabilities enabling the MnS consumer to express transient load requirements.
4.10.3 Potential solution
This solution proposes to reuse and enhance the existing RadioNetworkExpectation defined in 3GPP TS 28.312 [2].
Enhancement Aspect1: Add following attributes as the ExpectationTargets for the RadioNetworkExpectation to support transient overload requirements:
• PrbHighLoadRatio, it represents the load target for the radio network that the intent expectation is applied. The detailed definition for PrbHighLoadRatio see TS 28.554 [8]. The type is Real.
4.10.4 Evaluation of potential solutions
TBD
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.11 Use case #11: Enhancing intent feasibility check capability
| |
97c399698bf3bda50453286bc4472172 | 28.881 | 4.11.1 Description
| This use case describes proposed enhancements for feasibility check, enabling intent-driven MnS consumer to receive information regarding how to make an infeasible intent feasible together with feasibility check in case the intent is deemed as infeasible. This will support MnS consumers to understand the changes needed for a feasible intent.
In 3GPP TS 28.312 [1], clause 5.3.3.1 states the following:
“In case the result of intent fulfillment feasibility check is infeasible, MnS producer notifies the MnS consumer the reason of infeasibility and corresponding recommendations, then the MnS consumer decides how to handle the issue that intent is infeasible, e.g. update the intent, suspend the intent, delete the intent, etc.”
However, the issue is the IntentFeasibilityCheckReport specified in 3GPP TS 28.312 [1], clause 6.2.1.3.10 does not include corresponding recommendations. The MnS consumer may request exploration as specified in 3GPP TS 28.312 [1] Clause 5.3.5, based on inFeasibleExpectationInfos which is an optional attribute indicating the infeasible expectations including infeasible targets. However, if the inFeasibleExpectationInfos attribute is not supported, then it is very difficult for the intent-driven MnS consumer to update an infeasible intent as they do not have the recommendation on how to make it feasible.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.11.2 Potential requirements
| REQ-Intent_FEAS1: The intent driven MnS producer should have the capability to enable the authorized MnS consumer to receive the intent exploration report together with intent feasibility check report, if the feasibility check result is infeasible.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.11.3 Potential solutions
| |
97c399698bf3bda50453286bc4472172 | 28.881 | 4.11.3.1 Potential solution #1 for a new procedure of feasibility check with exploration
| This solution proposes to reuse the existing generic intent exploration report and intent feasibility check report defined in 3GPP TS 28.312 [1] and to enhance the Intent <<IOC>>.
In order to enable an MnS consumer to request a feasibility check with corresponding recommendations, a procedure is proposed with ensuring feasibility check with exploration is supported and enabled by a new allowed value “FEASIBILITYCHECK_WITH_EXPLORATION” that is to be included for intentMgmtPurpose attribute of Intent <<IOC>> which is specified in 3GPP TS 28.312, clause 6.2.1.4. When feasibility check with exploration is requested by the intent-driven MnS consumer, the intent driven MnS producer can report corresponding recommendations for infeasible expectations with existing IntentExplorationReport <<dataType>>, together with existing IntentFeasibilityCheckReport <<dataType>>.
Editor's note: This study may need to investigate how adding a new value to the intentMgmtPurpose attribute affects the state machine of the intent handling function.
4.11.4 Evaluation of potential solutions
TBD
4.12 Use case #12: Documentation for the overview of intent driven management functionalities
4.12.1 Description
In TS 28.312 [1], several intent management functionalities (including but not limited to Intent handling capability obtaining, intent exploration, intent feasibility check, intent report management) are introduced in different releases. However, the overview of intent driven management functionalities and corresponding usage for different phases are missing. It is important and useful to illustrate the intent driven management functionalities in TS 28.312 [1].
4.12.2 Potential requirements
4.12.3 Potential solution
It proposes to introduce a new concept section in clause 4 in TS 28.312 [1] as showed below to illustrate the overview of intent driven management functionalities.
4.X Overview of intent driven management functionalities for different phases
4.X.1 Introduction
The intent management functionalities for different phases include:
- Intent investigation and pre-evaluation
- Intent fulfilment
Above intent management functionalities applied for all intent categorizes, including Intent-CSC, Intent-CSP and Intent-NOP.
4.X.2 Intent investigation and pre-evaluation
In this phase, the MnS consumer finds out what intent content (a list of expectations) is feasible before expressing the intent expectations to be fulfilled by the intent handling function. The network (including NEs) will not be changed during this phase. This includes three functionalities:
- Intent handling capability obtaining. The functionality allows MnS consumer to find the intent handling function which has the necessary domain responsibilities and supports their required intent expectations. The MnS consumer can subscribe or query the intent handling capabilities (e.g., supported ExpectationObject, supportedExpectationTarget) for a specific intent handing function.
- Intent exploration. The functionality allows MnS Consumer to explore what the results of the wanted intent expectations would be and what is the best result the intent handling function can achieve. The MnS consumer can explore the best value(s) for specific intent target(s) (e.g. NumberofUEs, RANEnergyConsumption).
- Intent feasibility check. The functionality allows MnS consumer to verify or check the feasibility whether the proposed intent expectation is possible for an intent handling function.
4.X.3 Intent fulfilment:
In this phase, the intent handling function verifies the received intent and fulfils the given intent within its domains of responsibility. This includes following functionalities:
- Intent translation and execution. The functionality enables the intent handling function to translate the received intent to executable actions and performing the actions.
- Intent fulfilment evaluation. The functionality enables the intent handling function to evaluate the result about the intent fulfilment and generate the intent fulfilment reports. The intent handling function needs to continuously perform intent translation, execution and evaluation to ensure the intent fulfilment.
- Intent fulfilment negotiation. The functionality enables the intent handling function and the MnS consumer to agree on the best way to fulfil an intent.
- Intent conflict management. The functionalities enable the intent handling function to detect the intent related conflicts and collaborate with MnS consumer to resolve the conflicts. Following mechanism can be used to resolve the conflicts to improve the intent fulfilment:
- Intent priority and expectation preference. The functionality enable intent handling function resolve intent related conflicts based on intent priority and expectation preference specified by MnS consumer.
- Intent utility function. The functionality defines a method by which MnS consumers can express the relative value of an intent's expectations to assist the intent handling functions in fulfilling their intents in the most acceptable manner.
Editor’s Note: The above content maybe updated based on R20 progress.
4.12.4 Evaluation of potential solutions
TBD
4.13 Use case #13: Investigation on the intent lifecycle management state transition
4.13.1 Description
3GPP TS 28.312 [1] studies the state transitions during fulfilment phase. However, the intent Lifecycle management states that describes the pre-evaluation phase (i.e. intent feasibility check and intent exploration) and negotiation in intent fulfilment phase is missing. The main difference between the pre-evaluation phase, fulfilment phase and negotiation in intent fulfilment phase is on the intentMgmtPurpose. Thus, investigation on the intent lifecycle management state transition should be made, including the intent lifecycle management state transition diagram and the state transition events based on the intentMgmtPurpose.
4.13.2 Potential requirements
4.13.3 Potential solution
It proposes to add intent lifecycle management state transition diagram and intent lifecycle management state transition table to indicate the state transitions. Figure 4.13.2.1 shows the intent lifecycle management state diagram, where the number in the Figure identify the changes to the intent lifecycle management state. The explanations for the changes are described in Table 4.13.2.1.
Figure 4.13.2.1: intent lifecycle management state transition diagram
The transition numbers in the first column represent the intent lifecycle management state changes in Figure 4.13.2.1. The interactions specified under the column " Intent lifecycle management state transition events" of Table 4.13.2.1 shall be present for the transition.
NOTE 1: The transition numbers do not indicate any strict order and not correspond to procedure steps.
NOTE 2: Figure 4.13.2.1 may not be complete, and other states and transitions may be added.
Table 4.13.2.1: The intent lifecycle management state transition diagram table
Transition number
Intent lifecycle management state transition events
Intent lifecycle management state
1
The MnS producer creates the intent instances for intent exploration based on the received intent creation request, with the intentMgmtPurpose specified as "EXPLORATION" and the value of specific intent targets or contexts that need to be explored initially set to 'NULL'.
EXPLORATION
2
The MnS producer creates the intent instances for intent feasibility check based on the received intent creation request, with the intentMgmtPurpose specified as "FEASIBILITYCHECK".
FEASIBILITYCHECK
3
The MnS producer creates or modify the intent instances during fulfilment phase based on the received request, with the intentMgmtPurpose specified as " FULFILMENT_WITH_NEGOTIATION" or " FULFILMENT_ WITHOUT _NEGOTIATION ".
FULFILMENT
4
The MnS producer triggers the negotiation process. The process only can be triggered when intentMgmtPurpose is specified as " FULFILMENT_WITH_NEGOTIATION".
NEGOTIATION
5
The MnS producer finishes the negotiation process.
FULFILMENT
4.13.4 Evaluation of potential solutions
TBD
4.14 Use case #14: intent expectation satisfied information
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.14.1 Description
| In 3GPP TS 28.312 [1], the existing intentReport allows the IDMS consumer to obtain the values of KPI as indicated by the corresponding expectation targets. However, this information may not accurately reflect the actual situation. For example, considering an intent that includes a radio network expectation with the target aveDLRANUEthptTarget, the existing intent report provides the average value of the downlink RAN UE throughput and tells whether the intent expectation is fulfilled or not based on the average value. However, in the case where the average value is impacted by some extreme values, the fulfilled intent report will miss the information that the majority of UE experience doesn’t fulfil the UE throughput specified in the target. Such information matters because it helps the network operator to have a better understanding on the intent expectation satisfaction. The report can include the information besides the average information, which can be entity based, or time based, and is derived through statistical measurement or calculation, such as distributions, range or std deviation.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.14.2 Potential requirements
| REQ-Intent_IESI-1: The intent driven MnS producer should have a capability providing information besides average value in the intent report.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.14.3 Potential solutions
| TBD
4.13.4 Evaluation of potential solutions
TBD
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.15 Use case #15: Relation and Interaction with MnS producers for AI/ML Management
| |
97c399698bf3bda50453286bc4472172 | 28.881 | 4.15.1 Description
| This use case describes a scenario where an intent driven MnS producer interacts and coordinates with MnS producers for AI/ML management which is specified in 3GPP TS 28.105 [10]. Such coordination can enable the intent-driven MnS producer to both leverage AI/ML for its own internal intent handling tasks, as well as for when AI/ML processing might be needed in the network as an outcome of processing the intents.
As an example of leveraging AI/ML, in order for the intent-driven MnS producer to fulfil an expectation, the intent driven MnS producer may request to control the inference, e.g., activate/deactivate the ML model/models, configure the allowed ranges of the inference output parameters, request fine-tuning of an ML model etc. These are possible cases whereby the intent driven MnS producer can use an ML model and interact as an MnS consumer with MnS producers described in 3GPP TS 28.105 [10], clause 6 for ML model lifecycle management, such as ML model training, ML testing, ML model deployment, AI/ML inference, AIML Inference emulation.
In the case of AI/ML being leveraged as an outcome of the intent handling, an example can be considered whereby in order for the intent-driven MnS producer to check the feasibility of an intent or exploration of an intent in pre-evaluation phase as specified in 3GPP TS 28.312 [1], the intent-driven MnS producer may request, for example, AIML Inference emulation before the ML model is applied for fulfilment in production network.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.15.2 Potential requirements
| REQ-Intent_AIML-1: The intent driven MnS producer should have the capability to interact with MnS producers for ML model lifecycle management in both intent pre-evaluation phase and intent fulfilment phase.
|
97c399698bf3bda50453286bc4472172 | 28.881 | 4.15.3 Potential solutions
| TBD
4.15.4 Evaluation of potential solutions
TBD
4.16 Use case #16: Investigation on the applicability and potential impacts to support natural language intents translation
4.16.1 Description
In TR 28.914 [9], clause 5.6 described the use case for investigating potential impacts to support natural language intents translation. It also states that there are some use cases where the use of natural language for intent expression would be preferred. For example, on the human-machine interface presenting and expressing intent using natural language can lead to a more intuitive experience for human users. A concrete example of natural language RAN ES intent can be “Reduce 10% RAN energy consumption without impact on user experience for Region X”. The use case also described that if an intent interpreter is introduced to process intents described in natural language, it is necessary to clarify its position within the existing intent management framework, especially its relationship with the MnS consumer.
Scnenario#1: Intent Interpreter as a separate function outside the intent handling function. In this scenario, Intent interpreter receive the natural language intent and translate it into a formal intent (the formal intent contains the attributes and corresponding values to enable the intent handling function to decide the actions to fulfil the intent expectation) according to intent model defined in TS 28.312 [1] and consume the 3GPP Intent driven MnS with formal intent model provided by intent handling function.
Scnenario#2: Intent Interpreter is a function integrated in intent handling function. In this scenario, Intent Interpreter receives the natural language intent from consumer and translates it into detailed actions to be executed to fulfil the intent.
Based on above two deployment scenarios, this study can investigate whether there is a need for a potential solution to enhance the intent-driven management service to support natural language.
4.16.2 Potential requirements
TBD
4.16.3 Potential solution
TBD
4.16.4 Evaluation of potential solutions
TBD
|
97c399698bf3bda50453286bc4472172 | 28.881 | 5 Conclusions and Recommendations
| Editor's note: this clause will contain conclusions and recommendations for corresponding key issues identified in clause 4.
Annex A:
Change history
Change history
Date
Meeting
TDoc
CR
Rev
Cat
Subject/Comment
New version
2025-08
SA5#162
Initial version.
0.0.0
2025-09
SA5#162
S5-254121
draftTR
Draft TR v0.1.0
S5-254032 Add structure proposal
S5-254033 Add new issue for enhancement of radio service delivering and assurance scenarios
S5-254034 Add new issue for enhancement of radio network performance assurance scenarios
S5-254035 Add issue description, requirement and solution for intent decomposition
S5-254036 Add issue, requirements and potential solution for intent traceability
S5-254037 Invariant Guidance in Intent Contexts
Intent interpretation information
S5-254038 Add new issue for Intent exploration enhancement
S5-254039 Add Key Issue on Intent feasibility check
S5-254040 enhancement to support resource reservation
S5-254042 Add new issue for Intent handling capability registration and discovery
0.1.0
2025-10
SA5#163
S5‑254896
draftTR
Draft TR v0.2.0
S5-254645 Add solution for enhancement of radio network performance assurance scenarios
S5-254646 Add solution for enhancement of radio service delivering and assurance scenarios
S5-254647 Add new requirements and solution for enhancement radio service delivering and assurance scenarios
S5-254648 Add new use case for radio service assurance in transient overload scenarios
S5-254649 Add use-case description, requirement and solution for enhancing feasibility check
S5-254650 Add Solution to Support to Express Guarantee Requirements in an Intent
S5-254652 Add solution for Intent exploration enhancement
S5-254653 Documentation for the overview of intent driven management functionalities
S5-254654 Add use case for the investigation on the transition of intent Lifecycle management state
S5-254655 Add Use case on intent expectation satisfaction information
S5-254656 Add solution for Intent handling capability configuration, registration and discovery
S5-254657 Add use-case description, requirement and solution for relation and interaction with AIML
S5-254658 Add use case for the Investigation on the applicability and potential impacts to support natural language intents translation
S5-254659 Add solution for use case#3
S5-254660 Add potential solution for UC#4 intent traceability
S5-254884 Add description and requirements for intent guarantee UC#8
0.2.0
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 1 Scope
| This Technical Report (TR) explains how application (App) developers can leverage the services from enablement frameworks, namely Common API Framework (CAPIF), Edge Application Enablement (EDGEAPP), and the Service Enabler Architecture Layer (SEAL) to create their software.
To facilitate understanding, this study presents a set of representative deployment options. These are provided as examples to illustrate common operator scenarios, but they do not limit the options that the frameworks offer. The TR will cover the linkage between the available services with different vertical use cases, to guide App developers on how to use the different services to fulfil their use cases.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 2 References
| The following documents contain provisions which, through reference in this text, constitute provisions of the present document.
- References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific.
- For a specific reference, subsequent revisions do not apply.
- For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.
[1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications".
[2] 3GPP TS 23.222: "Common API Framework for 3GPP Northbound APIs (CAPIF); (Stage 2)".
[3] 3GPP TS 29.222: "Common API Framework for 3GPP Northbound APIs (CAPIF); (Stage 3)".
[4] 3GPP TS 23.558: "Architecture for enabling Edge Applications".
[5] 3GPP TS 29.558: "Enabling Edge Applications; Application Programming Interface (API) specification; (Stage 3)".
[6] 3GPP TS 23.434: "Service Enabler Architecture Layer for Verticals (SEAL); Functional architecture and information flows".
[7] 3GPP TS 33.122: "Security aspects of Common API Framework (CAPIF) for 3GPP northbound APIs".
[8] 3GPP TS 33.558: "Security aspects of enhancement of support for enabling edge applications".
[9] 3GPP TS 33.434: "Security aspects of Service Enabler Architecture Layer (SEAL) for verticals".
[10] IETF RFC 6749: "The OAuth 2.0 Authorization Framework".
[11] IETF RFC 7519: "JSON Web Token (JWT)".
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 3 Definitions of terms, symbols and abbreviations
| |
2b5c82519cec48b309e29881d9c51555 | 23.947 | 3.1 Terms
| For the purposes of the present document, the terms given in 3GPP TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in 3GPP TR 21.905 [1].
Application Client (AC): An application component, typically hosted on User Equipment (UE), that consumes services provided by an EAS (as defined in 3GPP TS 23.558 [4]).
Application Context: A set of data about the Application Client that resides in the Edge Application Server (as defined in 3GPP TS 23.558 [4]).
CAPIF API Invoker: An entity that consumes APIs exposed via CAPIF (as defined in 3GPP TS 23.222 [2]).
CAPIF API Provider: An entity that offers APIs via CAPIF (as defined in 3GPP TS 23.222 [2]).
Common API Framework (CAPIF): A 3GPP framework that provides common functions for API exposure, onboarding, discovery, and authorization across multiple domains and stakeholders (as defined in 3GPP TS 23.222 [2]).
Edge Application Server (EAS): An application component hosted on an edge node that provides services to ACs (as defined in 3GPP TS 23.558 [4]).
Edge Enabler Client (EEC): A client-side function, co-located with the AC, that interacts with the Edge Enabler Server (EES) for discovery, registration, and mobility support (as defined in 3GPP TS 23.558 [4]).
Edge Enabler Server (EES): A network function that provides EDGEAPP control plane services to EECs, including EAS discovery, mobility handling, and service continuity (as defined in 3GPP TS 23.558 [4]).
Edge Computing Service Provider (ECSP): An operator or provider responsible for deploying and operating EDGEAPP functions (ECS, EES, EAS) in the network (as defined in 3GPP TS 23.558 [4]).
Service Enabler Architecture Layer (SEAL): A framework providing reusable service enablers such as location, group communications, and messaging, available to vertical applications (as defined in 3GPP TS 23.434 [6]).
SEAL Server: A server implementing SEAL service enabler functions, possibly hosted centrally or at the edge, and accessible via CAPIF (as defined in 3GPP TS 23.434 [6]).
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 3.2 Symbols
| For the purposes of the present document, the following symbols apply:
<symbol> <Explanation>
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 3.3 Abbreviations
| For the purposes of the present document, the abbreviations given in 3GPP TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in 3GPP TR 21.905 [1].
AC Application Client
AEF API Exposing Function
AMF API Managing Function
APF API Publishing Function
CAPIF Common API Framework
CCF CAPIF Core Function
EAS Edge Application Server
ECSP Edge Computing Service Provider
EEC Edge Enabler Client
EES Edge Enabler Server
ECS Edge Configuration Server
EDGEAPP Edge Application Enablement
JSON JavaScript Object Notation
OAuth Open Authorization
REST Representational State Transfer
SEAL Service Enabler Architecture Layer
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4 Overview of Frameworks
| |
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4.1 Introduction
| Emerging 5G vertical services have requirements that can include low-latency, high-bandwidth, and context-aware processing at the network edge. To address these needs, 3GPP has developed several application enablement frameworks, each targeting a specific aspect of the application layer architecture:
- Common API Framework (CAPIF) – provides a unified, secure way to expose and consume service and network APIs across multiple domains and stakeholders, specified at stage-2 in 3GPP TS 23.222 [2]. CAPIF offers a harmonized approach for API development and exposure, allowing 3GPP network functions and even third parties to publish and discover APIs in a common environment.
- Edge Application Enablement (EDGEAPP) – specifies an architecture for applications at the network edge, including support for local service discovery, application lifecycle management, and continuity during user mobility, specified at stage-2 in 3GPP TS 23.558 [4]. EDGEAPP enables edge computing with minimal changes to existing applications, abstracting the edge infrastructure complexity behind standardized interfaces.
- Service Enabler Architecture Layer (SEAL) – defines a layer of reusable service enablers (e.g., location, group communications, messaging) available to vertical applications via standardized APIs, specified at stage-2 in 3GPP TS 23.434 [6]. By providing common functions that many industries require, SEAL allows different vertical applications to leverage the same core capabilities (for example, a group communication service or location service) instead of each vertical implementing its own.
This Technical Report provides an overview of how these frameworks work together and guides application developers on using them to build advanced services.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4.2 Common API Framework (CAPIF)
| This clause will provide the CAPIF role in secure API exposure, onboarding, and discovery.The Common API Framework (CAPIF) is a 3GPP framework that provides common functions for secure API exposure, discovery, onboarding, and authorization across multiple domains, 3GPP TS 23.222 [2]. Its primary role is to harmonize how service APIs are published and consumed, offering a single point of entry for applications (API invokers) towards the common API aspects (CAPF APIs) and a consistent method for API providers to expose their service APIs. By introducing a common API framework, CAPIF prevents fragmentation because application developers and third-party service providers can use one standardized approach, rather than dealing with independent interfaces for each service API. CAPIF is designed to be used in conjunction with other frameworks, where it is agnostic to the content of the APIs and instead focuses on the exposure mechanism for APIs.
CAPIF was first specified in 3GPP Release 15 and has since been adopted for exposing APIs from core network functions (e.g. NEF for network capabilities, MBMS, etc.) as well as from application-enablement layer services, in a secure and controlled manner.
The CAPIF architecture is based on a service-oriented model with three main entity groups as illustrated in Figure 4.2.1, specifically a CAPIF Core Function (CCF), the API Invoker and one or more API Provider domains:
- CAPIF Core Function (CCF): The CCF handles the onboarding of API invokers and API registering of API provider domain functions (establishing trust and identity for each). It maintains the service API registry for discovery, and provides common services like authentication, authorization token issuance and logging.
- API Exposing Function (AEF): A function on the API provider domain side that hosts one or more service APIs and exposes them to external consumers (API Invokers). The AEF is the entry point through which an API invoker accesses the provider’s service API; it enforces the service’s access policies and handles the runtime service API calls (e.g. rate limiting). A 3GPP network function or application server can presented themselves as an AEF to offer their service APIs in a standardized way.
- API Publishing Function (APF): A provider-side function responsible for publishing the service API information and endpoints of the AEF to the CAPIF Core Function. When an API is made available, the APF registers the API’s meta-information (e.g. name, version, resource paths, supported operations) with the CCF, so that prospective invokers can discover it.
- API Management Function: A provider-side management entity that allows the API provider to administer and monitor their APIs. The API Management Function interfaces with the CCF for reporting and retrieving logs of API invocations (for auditing), subscribing to events (e.g. notifications of certain API usage or threshold breaches).
- API Invoker: The consuming entity that uses CAPIF to discover and invoke service APIs. An API Invoker could be an external third-party application or another network function/application within the operator’s environment. The invoker first onboards with the CAPIF core function to establish its identity and credentials, then queries the API registry to discover available services, and finally invokes the desired API on the corresponding AEF, obtaining authorization from the CCF as required. In a developer context, the API Invoker role is played by the developer’s application or service that seeks to utilize operator or third-party APIs in a secure manner.
Figure 4.2.1: Functional model for the CAPIF as defined in 3GPP TS 23.222 [2]
CAPIF operates through a set of well-defined procedures and interfaces. At a high level, the lifecycle of API exposure and consumption in CAPIF is as follows:
1. Onboarding and registration: An API provider (such as an Edge Configuration Server in an edge computing scenario) will register with the CCF to become a recognized provider, typically using credentials or certificates issued by the operator’s CAPIF authority (see clause 8.28 of 3GPP TS 23.222 [2]). Similarly, an application acting as an API invoker (e.g. a vertical application server or a UE client) will onboard with the CCF to obtain a secure invoker identity (see clause 8.1 of 3GPP TS 23.222 [2]). The CAPIF core function can provision digital certificates or JSON Web Tokens (JWT) and record the entity’s public information (like its name, domain, and roles) in its database. This establishes a trust relationship: the CCF recognizes the entity as an authorized API provider or consumer in the system.
2. API Publication (see clause 8.3 of 3GPP TS 23.222 [2]): Once registered, an API provider exposes its service API through CAPIF. The provider’s API Publishing Function sends the API description to the CAPIF core (via the CAPIF publishing interface, e.g. CAPIF-4) to register the API in the registry. This publication typically includes the API’s endpoint URI (which can point to the AEF), the resource structure and methods, required security scopes or policies, and any other metadata (version, category, etc.). For instance, in an EDGEAPP scenario, after an Edge Configuration Server (ECS) registers, it would publish its available service APIs (like edge node registration API, configuration APIs) to CAPIF so that edge clients or other components can later discover them. The CAPIF core can verify that the publishing entity is authorized to expose those APIs (e.g. checking that the provider is indeed entitled to publish an “Edge Configuration” service). Once published, the API is discoverable by the authorized invokers.
3. Discovery (see clause 8.7 of 3GPP TS 23.222 [2]): An API invoker queries the CAPIF Core Function to discover information a service API. Discovery can be performed using various criteria (service name, category, provider, etc.). The CCF will return the details of the AEF address, interface details and information on how to invoke it. CAPIF supports a federated discovery if multiple CCFs are interconnected, but within a single domain the CCF acts as the central lookup for all APIs. For example, an Edge Enabler Client (EEC) that needs to call a service on an Edge Application Server (EAS) can query CAPIF (via the CCF) to get the API endpoint and requirements for the EAS’s service API.
4. API Invocation with Authorization: Before invoking the target API, the API invoker obtains authorization from CAPIF. In CAPIF’s security framework, the CCF can function as an OAuth2 Authorization Server for API access tokens, see 3GPP TS 33.122 [7]. The invoker (acting as an OAuth client) requests an access token from the CCF, proving its identity (using credentials established at onboarding) and specifying the API it wants to call. The CCF issues a OAuth2 token if the invoker is allowed to use that API, with appropriate scope/permissions encoded. Armed with this token, the invoker then sends the API request to the provider’s AEF (over the CAPIF-internal interface to the AEF, e.g. CAPIF-2) and includes the token in the invocation. The AEF validates the token (by verifying its signature and checking with CCF if needed) and, if valid, allows the request to be processed by the underlying service. This ensures that only properly onboarded and authorized applications can invoke protected APIs. CAPIF also supports additional authorization models – for example, a mode where a “Resource Owner” (end-user or owner of the accessed resource) involvement is required for authorization (termed Resource Owner Aware API Access, RNAA).
5. Logging and Monitoring (see clause 8.19 and 8.21 respectively of 3GPP TS 23.222 [2]): As APIs are invoked, CAPIF provides a logging service. Each service API invocation and response can be logged via the CAPIF core (or the provider’s API Management Function reporting to CCF), recording details such as who invoked which API, when, and with what result. These logs enable auditing and troubleshooting. CAPIF can also emit events or notifications – for instance, if an API provider wants to be notified when a new invoker enrols to its API or when certain thresholds are exceeded, the framework can support that through an events interface.
For application developers, CAPIF greatly simplifies integration with 3GPP network services and other service platforms. Instead of dealing with separate authentication and discovery mechanisms for each API (e.g. one for location services, another for messaging, etc.), a developer can rely on CAPIF to provide a one-stop hub for all API interactions. For example, a developer building a fleet management application could use CAPIF to discover the operator’s “vehicle location API” and invoke it securely, without needing to know the internal details of the network’s location function – CAPIF mediates trust and access. Likewise, if the developer’s application provides its own API (say it’s an analytics service that the operator might call), the developer can publish this API via CAPIF so that it’s uniformly accessible to other authorized parties.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4.3 Edge Application Enablement (EDGEAPP)
| This clause will provide the EDGEAPP role in application lifecycle and orchestration at the edge.Edge Application Enablement (EDGEAPP) has been specified by 3GPP to enable applications to be deployed and managed at the edge of the network (closer to end users) to achieve ultra-low latency and high bandwidth efficiency, 3GPP TS 23.558 [4]. By offloading computation and services from centralized cloud to edge nodes (e.g. servers co-located with base stations or local data centers), EDGEAPP enables use cases like real-time video analytics, AR/VR, V2X, and industrial automation that require immediate processing and minimal backhaul delay. The key goal of EDGEAPP is to make this edge deployment transparent to application developers – applications should not need major redesign to benefit from edge computing. The framework accomplishes this by providing a standardized set of functions that handle discovery of edge services, provisioning and relocation of application instances, and continuity of sessions as users move between edge sites. In essence, EDGEAPP extends the 3GPP system with an application-layer control plane for edge computing, while hiding network complexity from the application itself.
The EDGEAPP architecture introduces several new functional entities (as depicted in Figure 4.3-1), which can be grouped into client-side and network-side components:
- Application Client (AC): The end-user application component that consumes a service. In an edge computing scenario, this could be, for example, a video streaming app, a game client, or an IoT application running on a UE (User Equipment). The AC is typically developed by the Application Service Provider and contains the core application logic for the end user. Importantly, the AC itself does not directly interface with the edge network functions; instead, it interacts with the Edge Enabler Client via local APIs on the device. This design keeps the AC simple and focused on the service logic, while offloading networking tasks to the Edge Enabler Client.
- Edge Enabler Client (EEC): A client-side middleware function residing on the UE (or whatever device hosts the AC). The EEC is the bridge between the application client and the edge system. It exposes a set of APIs (referred to as EDGE-5 interfaces in 3GPP) to the AC, through which the AC can request edge services (such as registering to the edge system, discovering available edge application servers, etc.). Internally, the EEC communicates with the network’s edge enablement functions: it registers the device/application with an Edge Enabler Server, keeps track of the current “serving” edge zone, and orchestrates any changes needed when the user moves. From the developer’s perspective, the EEC is crucial – by calling the EEC’s APIs, the developer’s application can utilize edge capabilities without needing to implement complex protocols. The EEC also handles obtaining any needed network authorization (for example, via CAPIF tokens when calling edge services) on behalf of the AC.
- Edge Enabler Server (EES): A network-side function that is typically deployed at an edge data center or site serving a particular geographical zone. The EES is the core control-plane entity in the EDGEAPP framework. It manages edge application services within its zone and interfaces with EECs on UEs. Key responsibilities of the EES include registering new Edge Application Servers (EAS) when they become available in its zone, handling UE registrations (via EECs) so that the system knows which UEs are using edge services in that area and facilitating service discovery for UEs (telling an EEC which local EAS can provide the service the AC needs). The EES also monitors UE mobility (through integration with the cellular network or via the EEC’s updates) – when a UE is about to move or has moved to another zone, the EES can trigger Application Context Relocation (ACR) procedures to ensure continuity of the service session. In summary, the EES is analogous to a local edge broker/manager that coordinates between UEs and application servers at the edge.
- Edge Application Server (EAS): The actual application server instance that runs on the edge node, providing the service logic to be consumed by the AC. This could be, for example, an instance of a video analytics engine, a game server, or a data caching server, deployed in the edge cloud. The EAS is owned by the application provider (ASP) and hosts the service that the AC uses. Each EAS registers with the local EES so that the system knows the service it offers and can advertise it to interested clients. The EAS relies on the EES for certain lifecycle operations; for instance, the EES can instruct the EAS to transfer state or shut down when a user leaves the zone. The EAS can also consume network and edge services itself – for example, an EAS might call an mobile network operator provided API (like a location service via CAPIF) to augment its application or use EES-provided APIs for retrieving information about UEs or network conditions. In EDGEAPP terminology, there are reference points (like EDGE-3 and EDGE-7) for EAS to interact with the EES and potentially with operator network exposures.
- Edge Configuration Server (ECS): A centralized function that oversees the configuration and coordination of the edge system across multiple edge sites. The ECS can be thought of as the master controller in a multi-edge deployment. It maintains a global view of all edge sites (each with an EES and perhaps multiple EAS instances). The ECS provides services such as: answering queries from EECs or EESs about where certain application services are available (e.g., if a UE in zone A needs a service that’s currently only in zone B, the ECS can help route or instantiate it), bootstrapping new edge nodes (EESs) into the system, and storing subscription information or policies that apply network wide. In some scenarios, the EEC can contact the ECS at initial registration (EDGE-4 interface) to obtain information like the address of its serving EES if the UE is not preconfigured. The EESs also typically interface with the ECS (EDGE-6) to report new EAS deployments or to get updates on global configurations. The ECS, being a central point, can use CAPIF/NEF to interface with core network functions – for example, to fetch network resource exposure (like cell congestion status) or to publish its own APIs for other systems. The Edge Computing Service Provider (ECSP), for example the mobile network operator, is responsible for running the ECS, EES, and providing the infrastructure on which EAS instances run.
Figure 4.3-1: EDGEAPP Architecture, as specified in 3GPP TS 23.558 [4]
The typical sequence of interactions in EDGEAPP to deliver an edge-augmented service is outlined below (note that some steps can vary depending on deployment options, but this reflects a common scenario in a single operator’s network):
1. Edge Infrastructure Registration: Before any end-user device participates, the edge nodes must be set up. Each Edge Enabler Server (EES) that comes online (for a new edge site) registers with the Edge Configuration Server (ECS). The ECS thus maintains a registry of all EESs and their capabilities/zones. This registration can involve mutual authentication and authorization – often incorporating CAPIF if the ECS’s APIs are published via CAPIF. In our context, the operator ensures all EES instances are known to the ECS (either through provisioning or dynamic discovery in larger deployments). Similarly, when an application provider wants to deploy an Edge Application Server (EAS) in a zone, they might coordinate with the operator such that the EAS gets registered with the local EES. The EES will record what service the EAS offers (e.g. “VideoAnalyticsService v1”) and possibly pass that information up to the ECS for global awareness. At this stage, CAPIF can be used: for example, the ECS and EES can onboard to CAPIF as API providers and invokers so that their APIs (like “register EAS” or “query edge info”) are accessible securely. By the end of this step, the edge control-plane is prepared: EES know about local EAS services, and the ECS knows about all EESs (and possibly their EAS listings).
2. UE Discovery of Edge Service (Registration): When a user’s device (UE) powers on or an application starts, the Edge Enabler Client on the UE needs to discover which edge it should connect to. There are multiple ways this can happen. The UE could have a pre-provisioned address for an EES (perhaps provided by the operator’s subscription or via a device management system); or the UE could perform a DNS query or receive an indication from the network (e.g., during attach, the network might provide an “edge pointer”). Another method is the bootstrap via ECS: the EEC contacts the central ECS to ask, “where is my serving EES?” and the ECS responds with the address of the optimal EES for that UE (based on the UE’s location or subscription). Once the EEC knows which Edge Enabler Server to use, it proceeds to register with that EES. EEC-EES registration involves the EEC introducing the AC (application client) and itself to the EES: essentially saying “this UE/application wants to use edge services in your zone.” The EES authenticates the EEC (possibly via network credentials or certificates; security can rely on 3GPP credentials or CAPIF tokens if the EEC is acting as an invoker) and then creates a context for that AC on the edge. After a successful registration, the EES knows this UE (often by a unique Edge UE ID) and can start providing it with services like discovery notifications. The EEC will typically get a confirmation that it is registered. From the AC developer’s viewpoint, this whole discovery and registration process is triggered by a simple API call like Eeec_ACRegistration() from the AC to the EEC – the EEC and the network handle the rest (including contacting ECS or using CAPIF as needed).
3. Service Discovery and Consumption: After registration, the AC can request an actual application service at the edge. Suppose the AC needs a “Video Analytics” service. The AC (through an API call to EEC, e.g. Eeec_EASDiscovery(serviceType)) asks the EEC to find an appropriate Edge Application Server (EAS) that provides that service. The EES receives this request via the EEC and checks whether an EAS in its zone offers the requested service. If yes, it selects the optimal EAS (there can be multiple instances or versions) and returns the details (e.g. the EAS’s IP address or domain and any required info to connect) to the EEC, which then passes it to the AC. If no local EAS can serve it, the EES might query the ECS to see if the service is available in another zone or if a new instance should be instantiated. Assuming an EAS is found in the local zone (Edge Zone 1 for example), the AC can then start communicating with that EAS to consume the service. Typically, the AC’s requests will still go through the EEC as a proxy (the exact data plane path can vary it could be AC → EEC (on device) → direct IP to EAS, or AC → EEC → EES → EAS, depending on whether the EES needs to anchor the traffic). In any case, the AC’s traffic to the EAS is application-layer communication (e.g. HTTP requests, video stream) and the EES/EEC ensure it is routed to the edge rather than a distant server.
4. It’s worth noting the security aspect here: if the EAS’s API is protected (which would be the usual scenario expected), the AC/EEC can need to obtain an authorization token to invoke it. This is where CAPIF integration comes in – the EAS can expose its service API via CAPIF (with the EAS acting as an API provider and the EEC or AC as the invoker). In practice, the EEC might use CAPIF to get a token for the EAS’s API and include it in AC’s requests. However, these steps are typically hidden from the developer; they happen under the hood. The developer just sees that after discovery; the AC can call the EAS (likely the EEC provides a local API call that internally forwards to the EAS). In summary, the AC starts consuming the edge service, enjoying much lower latency because the EAS is nearby and potentially tailored to local conditions.
5. Mobility and Service Continuity: A key feature of EDGEAPP is handling what happens when a user moves out of the coverage of one edge site to another (e.g., driving from one city to the next). Without EDGEAPP, the session to the first edge server would break or see a drastic performance drop. EDGEAPP’s design allows for Application Context Relocation (ACR) to preserve service continuity. In our example, as the UE (with AC and EEC) moves from Edge Zone 1 to Edge Zone 2, the system must transfer the ongoing session from the current EAS (in Zone 1) to a new EAS in Zone 2 (that provides the same service). The EES in Zone 1, detecting that the UE is leaving (possibly via network triggers or the UE’s EEC notifying loss of connection), coordinates with the EES in Zone 2 and the central ECS to set up a target EAS. If a suitable EAS is already running in Zone 2, it can prepare to take over the session; if not, the operator might instantiate one on the fly (depending on how dynamic the system is – this is more advanced). The state of the application (the “application context”) can be transferred from EAS1 to EAS2 – this could involve state sync or handing off relevant data (the exact method can depend on application design; EDGEAPP can signal the EAS to push state). Once Zone 2’s EAS is ready, the EES2 (Edge Enabler Server in Zone 2) will finalize the relocation: the EEC on the UE is informed of the new EAS endpoint (via a notification or updated discovery result). The AC then seamlessly continues communication with EAS2. From the AC’s viewpoint, it might just receive an event or callback that a new server endpoint is now in use – the content of the service continues with minimal disruption. The AC does not need to re-register or rediscover manually; the EEC and EES handle that transition. EDGEAPP thus ensures that mobility across edges is managed gracefully, which is critical for use cases like vehicular connectivity or users walking across a city with an AR application. This continuity is achieved while still enforcing security (if a new token is needed for the new EAS, the EEC would obtain it via CAPIF as well, transparently to the AC).
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4.4 Service Enabler Architecture Layer (SEAL)
| This clause will provide the SEAL role in the provision of reusable service enablers for vertical applications.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 4.4 Inter-framework Relationships Overview
| This clause will describe the role of each framework can provide in an overall framework.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5 Use Cases
| Each use case is expected to include end-to-end service flow and infrastructure details (e.g., number and placement of edge nodes, optional use of SEAL services for a single PLMN domain)
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1 Crowd Counting Video Analytics
| |
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.1 Introduction
| This section describes a Crowd Counting Video Analytics use case. The use case has been selected to illustrate how application developers can leverage the EDGEAPP framework to discover and utilize Edge infrastructure that supports low-latency video analytics. Specifically, it demonstrates the benefits of deploying computation close to the user equipment (UE) when processing video streams in real time.
For EDGEAPP infrastructure discovery, the CAPIF framework is leveraged to demonstrate how CAPIF exposes standardized API information that enables developers to optimize their applications for efficient utilization of edge services. Through CAPIF, an API Catalogue is made available across all Edge Zones, allowing applications to discover available zones and obtain the corresponding API endpoints required to access and consume the Video Analytics services.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.2 Use case Description
| “Acme” company is developing a crowd counting video analytics application that leverages edge computing to process video in real time. The application consists of an Application Client (AC) running on a User Equipment (UE), which captures live video and streams it for analytics, and a video analytics application running on an Edge Application Server (EAS), which performs the crowd counting function.
The performance of the video analytics function depends on latency and network efficiency. Deploying the analytics server closer to the UE can significantly reduce latency and minimize the impact of transmitting large video streams across the network.
Developers can choose between two deployment models:
- Centralized Deployment: The video analytics server is deployed in a central data center, requiring all UEs to stream video to the central location.
- Edge Deployment: The video analytics server is deployed within Edge Zones, allowing video streams to be processed locally, thereby reducing backhaul traffic and latency.
The UE also hosts an Edge Enabler Client (EEC). The AC communicates with the EEC via the EDGE-5 reference point. The EEC hides all network complexity and interacts with the Edge Enabler Server (EES), the Edge Configuration Server (ECS), and the CAPIF Core Function (CCF) to enable the AC to consume the video analytics service.
As depicted in Figure 5.1.1-1, the topology considered for this use case includes:
• one CCF,
• one ECS,
• two EES (each serving one edge zone),
• two EAS (one per edge zone),
• one UE with EEC and AC.
Figure 5.1.1-1: Crowd Counting Video Analytics use case topology
The UE starts attached in Edge Zone 1, where the AC streams video to EAS 1. When the UE moves into Edge Zone 2, the system triggers an Application Context Relocation (ACR) to transfer the session to EAS 2, preserving service continuity with minimal disruption. The AC itself is only aware that the EEC provides it with an updated service endpoint; all control-plane interactions are performed by the EEC.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.3 Use case Realisation over Application Enablement Frameworks
| |
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.3.1 General
| From the developer’s point of view, it is essential to understand which entity handles which tasks in the EDGEAPP system. The AC is deliberately kept simple: it talks only to the EEC, and all network complexity is abstracted away. When the AC is designed, only the calls to the EEC’s APIs need to be implemented to leverage the EDGEAPP capabilities. Everything else — serving EES discovery, ECS interaction, and service API resolution via CAPIF — is handled transparently by the edge system. Section 5.1.3.2, though, explain the end-to-end flows for developers to understand the full system.
For this use case, we are considering centralized CAPIF core function deployment option as described in Annex A.5.3 of 3GPP TS 23.558 [4].
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.3.2 AC Discovering the Serving EES
| The EEC running in the UE, needs to discover its EES. The EEC can learn which EES to register with through multiple mechanisms:
• Pre-provisioned in the UE/application profile.
• DNS-based discovery, using operator domain naming conventions.
• Network-provisioned, delivered at onboarding or during mobile attachment.
• Bootstrap via ECS, where the EEC queries ECS to obtain the correct serving EES.
|
2b5c82519cec48b309e29881d9c51555 | 23.947 | 5.1.3.3 CAPIF roles of EDGEAPP entities
| While AC does not interact directly with CCF, it is important to understand the roles that each EDGEAPP entity plays in relation to CAPIF to consume APIs. In this use case, the EEC behaves as a CAPIF API Invoker. The EEC needs to discover EAS APIs and obtain OAuth tokens to consume them. In particular, the EEC queries the CCF to discover the Video Analytics service API exposed by the EAS.
The EEC then exposes the resolved API base URI and parameters to the AC via EDGE-5 (Eeec_EASDiscovery). The AC uses this information to send its video stream (i.e., application data traffic) directly to the EAS and expects to receive analytics results based on that. EES acts as an API Invoker consuming APIs from ECS and is an API provider as it exposes EES APIs to ECS. EAS acts as an API Invoker consuming APIs from EES and is an API provider as it exposes EAS APIs plus Application specific APIs. ECS acts as API Provider as it exposes APIs to EES.
With these considerations, the next section provides an end-to-end view with step-by-step flows describing all EDGEAPP and CAPIF entities and how they interact based in the Developers´ use of EDGE-5 APIs between AC and EEC.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.