hash stringlengths 32 32 | doc_id stringlengths 5 12 | section stringlengths 5 1.47k | content stringlengths 0 6.67M |
|---|---|---|---|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6 Solutions
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.0 Mapping of solutions to key issues
| Table 6.0-1 Mapping of solutions to key issues
KI #1
KI #2
KI #3
KI #4
KI#5
KI#6
Sol #1
X
Sol #2
X
X
Sol #A3
X
Sol #4
Sol #x5
X
X
Sol #6
Sol #7
Sol #Y18
X
Sol #Y9
X
Sol #10
6.X21 Solution #X21: New Service on Energy Saving Assistance
6.X21.1 Solution Description
6.X21.1.0 General
The following clauses specify the procedure and information flow on energy saving assistance provisioning.
6.X21.1.1 Procedure
Figure 6.X21.1.1-1: Energy saving assistance
1. A consumer (e.g., VAL server, SEAL server/client) sends energy saving assistance request to SEAL server. The request includes information as described in Table 6.X21.1.2.2-1.
2. The SEAL server authenticates and authorizes the request. If authorized, the SEAL server determines the next actions (e.g. collect energy-related data, monitor energy-related information/status) for assisting energy saving. The ESE server can consolidate similar requests from Consumers (e.g., data collection from same source into single southbound requests, avoid re-fetching data that it already has if the request parameters allow it such as time frame matches, etc.).
3. The SEAL server sends energy-related data collection response to the consumer. The response includes information as described in Table 6.X21.1.2.3-1.
4. The SEAL server collects energy consumption information, monitor energy status, and generates assistance information for energy saving based on the data/information obtained in step 4.
5. The SEAL server sends energy saving assistance notify to the consumer with the assistance information. The notify includes information as described in Table 6.X21.1.2.4-1.
Editor’s Note: The update of the solution is FFS.
6.X21.1.2 Information Flows
6.X21.1.2.1 General
The following information flows are specified for energy saving assistance.
6.X21.1.2.2 Energy saving assistance request
Table 6.X21.1.2.2-1 shows the request sent by a SEAL service consumer to a SEAL server for energy saving assistance procedure.
Table 6.X21.1.2.2-1: Energy saving assistance request
Information element
Status
Description
Requestor identifier
M
The identifier of the requestor.
VAL service ID
M
The identity of the VAL service for which the request applies.
Target energy saving entity(ies)
M
The identity(ies) of the target energy saving entity(ies) (e.g. VAL UE ID(s), SEAL server ID(s))
Assistance requirements
M
Identifies the requirement for the assistance on energy saving.
>Assistance information type
M
Identifies the assistance information type required for energy saving (e.g. information on entity which consume the most/least energy, assistance information for selection of entity(ies)).
6.X21.1.2.3 Energy saving assistance response
Table 6.X21.1.2.3-1 shows the response sent by the SEAL server to the consumer for energy saving assistance procedure.
Table 6.X21.1.2.3-1: Energy saving assistance response
Information element
Status
Description
Result
M
The result of the request (positive or negative acknowledgement).
Subscription ID
O
Identifier of the subscription. It shall be provided if the request if the Result is positive acknowledgement.
6.X21.1.2.4 Energy saving assistance notify
Table 6.X21.1.2.4-1 shows the notification sent by the SEAL server to the consumer for energy saving assistance procedure.
Table 6.X21.1.2.4-1: Energy saving assistance notify
Information element
Status
Description
Success status
O
(NOTE)
Indicates that energy saving assistance was successful.
>Assistance information
M
Provides energy saving assistance information corresponding to the request.
Failure response
O
(NOTE)
Indicates that energy saving assistance was failure.
>Cause
M
Reason for the failure.
NOTE: One of the information elements shall be provided in the output.
6.X21.2 Architecture impacts
Editor’s Note: The architecture impacts of the solution is FFS.
6.X21.3 Solution evaluation
Editor’s Note: The evaluation of the solution is FFS.
6.x2 Solution #x2: Edge Application Server Discovery by Considering Renewable energy
6.x2.1 Solution Description
The proposed solution solves the problem listed in KI#1 and KI#3 and proposes to consider Renewable energy in edge computing.
In Edge Computing deployment, an application service may be served by multiple Edge Application Servers typically deployed in different sites. These multiple Edge Application Servers that host service may use a single IP address (anycast address) or different IP addresses. To start such a service, the UE needs to know the IP address(es) of the Application Server(s) serving the service. The UE may do a discovery to get the IP address(es) of a suitable Edge Application Server (e.g. the closest one), so that the traffic can be locally routed to the Edge Application Server and service latency, traffic routing path and user service experience can be optimized.
During the EAS discovery procedure, that the renewable energy of EAS is not considered. In order to save the energy consumption and reduce the carbon emission, the operators may consider using renewable energy or green energy to replace traditional energy.
6.x2.2 Architecture impacts
In this solution, it doesn’t change the overall architecture of enabling edge applications defined in section 6.2 of 3GPP TS 23.558 [y8]. But the following enhancements are added:
- The EES can subscribe to EAS whether renewable is consumed by EAS. If the EAS consumes the renewable energy, for example in the morning time that solar energy can be consumed, the EES may select this EAS to save the energy.
6.x2.3 Procedure
6.x2.3.1 Procedure of subscription Energy information from EAS
Figure 6.x2.3.1-1: EAS energy information subscription
1. The EES sends an EAS energy information subscription request to the EAS. The EAS energy information subscription request includes the EES ID along with the security credentials, Event ID and time period to subscribe to information about energy information of EAS.
The energy information includes Renewable energy of EAS. The Renewable energy of EAS includes that either the indication of the EAS consumes renewable energy or not, or the Renewable energy factor of EAS.
The time period indicates the EAS to notify the Energy information of EAS in the certain time period in the future.
2. Upon receiving the request from the EES, the EAS checks if the EEC is authorized to subscribe for information of the requested EAS(s).
3. If the processing of the request was successful, the EAS sends an EAS energy information subscription response to the EES, which includes the subscription identifier and may include the expiration time, indicating when the subscription will automatically expire.
4. The EAS sends an EAS renewable energy utilization notification to the EES.
Editor’s nNote: How EAS know the capability of renewable energy supporting is FFS.
6.x2.3.2 Procedure of EAS discovery considering renewable energy information
The procedure listed here are based on the procedure defined in section 8.5.2.2 of 3GPP TS 23.558 [y8].
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1 Solution Description
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.0 General
| The following clauses specify the procedure and information flow on energy saving assistance provisioning.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.1 Procedure
|
Figure 6.1.1.1-1: Energy saving assistance
1. A consumer (e.g., VAL server, SEAL server/client) sends energy saving assistance request to SEAL server. The request includes information as described in Table 6.1.1.2.2-1.2. The SEAL server authenticates and authorizes the request. If authorized, the SEAL server determines the next actions (e.g. collect energy-related data, monitor energy-related information/status) for assisting energy saving. The ESE server can consolidate similar requests from Consumers (e.g., data collection from same source into single southbound requests, avoid re-fetching data that it already has if the request parameters allow it such as time frame matches, etc.).
3. The SEAL server sends energy-related data collection response to the consumer. The response includes information as described in Table 6.1.1.2.3-1.4. The SEAL server collects energy consumption information, monitor energy status, and generates assistance information for energy saving based on the data/information obtained in step 4.
5. The SEAL server sends energy saving assistance notify to the consumer with the assistance information. The notify includes information as described in Table 6.1.1.2.4-1.
Editor’s Note: The update of the solution is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.2 Information Flows
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.2.1 General
| The following information flows are specified for energy saving assistance.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.2.2 Energy saving assistance request
| Table 6.1.1.2.2-1 shows the request sent by a SEAL service consumer to a SEAL server for energy saving assistance procedure.
Table 6.1.1.2.2-1: Energy saving assistance request
Information element
Status
Description
Requestor identifier
M
The identifier of the requestor.
VAL service ID
M
The identity of the VAL service for which the request applies.
Target energy saving entity(ies)
M
The identity(ies) of the target energy saving entity(ies) (e.g. VAL UE ID(s), SEAL server ID(s))
Assistance requirements
M
Identifies the requirement for the assistance on energy saving.
>Assistance information type
M
Identifies the assistance information type required for energy saving (e.g. information on entity which consume the most/least energy, assistance information for selection of entity(ies)).
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.2.3 Energy saving assistance response
| Table 6.1.1.2.3-1 shows the response sent by the SEAL server to the consumer for energy saving assistance procedure.
Table 6.1.1.2.3-1: Energy saving assistance response
Information element
Status
Description
Result
M
The result of the request (positive or negative acknowledgement).
Subscription ID
O
Identifier of the subscription. It shall be provided if the request if the Result is positive acknowledgement.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.1.2.4 Energy saving assistance notify
| Table 6.1.1.2.4-1 shows the notification sent by the SEAL server to the consumer for energy saving assistance procedure.
Table 6.1.1.2.4-1: Energy saving assistance notify
Information element
Status
Description
Success status
O
(NOTE)
Indicates that energy saving assistance was successful.
>Assistance information
M
Provides energy saving assistance information corresponding to the request.
Failure response
O
(NOTE)
Indicates that energy saving assistance was failure.
>Cause
M
Reason for the failure.
NOTE: One of the information elements shall be provided in the output.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.2 Architecture impacts
| Editor’s Note: The architecture impacts of the solution is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.1.3 Solution evaluation
| Editor’s Note: The evaluation of the solution is FFS.
6.2 Solution #2: Edge Application Server Discovery by Considering Renewable energy
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.1 Solution Description
| The proposed solution solves the problem listed in KI#1 and KI#3 and proposes to consider Renewable energy in edge computing.
In Edge Computing deployment, an application service may be served by multiple Edge Application Servers typically deployed in different sites. These multiple Edge Application Servers that host service may use a single IP address (anycast address) or different IP addresses. To start such a service, the UE needs to know the IP address(es) of the Application Server(s) serving the service. The UE may do a discovery to get the IP address(es) of a suitable Edge Application Server (e.g. the closest one), so that the traffic can be locally routed to the Edge Application Server and service latency, traffic routing path and user service experience can be optimized.
During the EAS discovery procedure, that the renewable energy of EAS is not considered. In order to save the energy consumption and reduce the carbon emission, the operators may consider using renewable energy or green energy to replace traditional energy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.2 Architecture impacts
| In this solution, it doesn’t change the overall architecture of enabling edge applications defined in section 6.2 of 3GPP TS 23.558 [8]. But the following enhancements are added:
- The EES can subscribe to EAS whether renewable is consumed by EAS. If the EAS consumes the renewable energy, for example in the morning time that solar energy can be consumed, the EES may select this EAS to save the energy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.3 Procedure
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.3.1 Procedure of subscription Energy information from EAS
|
Figure 6.2.3.1-1: EAS energy information subscription
1. The EES sends an EAS energy information subscription request to the EAS. The EAS energy information subscription request includes the EES ID along with the security credentials, Event ID and time period to subscribe to information about energy information of EAS.
The energy information includes Renewable energy of EAS. The Renewable energy of EAS includes that either the indication of the EAS consumes renewable energy or not, or the Renewable energy factor of EAS.
The time period indicates the EAS to notify the Energy information of EAS in the certain time period in the future.
2. Upon receiving the request from the EES, the EAS checks if the EEC is authorized to subscribe for information of the requested EAS(s).
3. If the processing of the request was successful, the EAS sends an EAS energy information subscription response to the EES, which includes the subscription identifier and may include the expiration time, indicating when the subscription will automatically expire.
4. The EAS sends an EAS renewable energy utilization notification to the EES.
Editor’s Note: How EAS know the capability of renewable energy supporting is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.3.2 Procedure of EAS discovery considering renewable energy information
| The procedure listed here are based on the procedure defined in section 8.5.2.2 of 3GPP TS 23.558 [8].
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.5.2.2 Request-response model
| Pre-conditions:
1. The EEC has received information (e.g. URI, IP address) related to the EES;
2. The EEC has received appropriate security credentials authorizing it to communicate with the EES as specified in clause 8.11; and
3. The EES is configured with ECSP's policy for EAS discovery.
NOTE 1: Details of ECSP's policy are out of scope.
Figure 8.5.2.2-1: EAS Discovery procedure
1. The EEC sends an EAS discovery request to the EES. The EAS discovery request includes the requestor identifier [EECID] along with the security credentials and may include EAS discovery filters, EEC service continuity support, and may also include UE location to retrieve information about particular EAS(s) or a category of EASs, e.g. gaming applications, or Edge Applications Server(s) available in certain service areas, e.g. available on a UE's predicted or expected route. The request may include an EAS selection request indicator. If an e2e tunnel is used in the UE for applications (e.g. user configured tunnel service), and if the tunnel service provider and the ECSP are from the same organization and only if user grants the permission, then the EEC provides the tunnel information for the associated applications in the request to the EES considering user consent. The EAS discovery filter may include energy requirements to the EAS, e.g., Energy mode, Energy type supports. The Energy type is about requirement to the EAS on its supported energy type (e.g. non-renewable energy and/or renewable energy).
NOTE 2: The e2e tunnel case is limited to case where the tunnel service is deployed in home domain in the present release.
2. Upon receiving the request from the EEC, the EES checks if the EEC is authorized to discover the requested EAS(s). The authorization check may apply to an individual EAS, a category of EASs or to the EDN, i.e. to all the EASs. If EAS discovery filters contain energy requirements, the EES checks whether the EAS satisfy the requirements on Energy type supports in the EAS discovery requests. If the EAS fulfil the energy requirements, then EES adds the identified EAS(s) into the list of discovered EAS(s). If UE's location information is not already available, the EES obtains the UE location by utilizing the capabilities of the 3GPP core network as specified in clause 8.10.3. If EAS discovery filters are provided by the EEC, but it does not contain Application group profile, the EES identifies the EAS(s) based on the provided EAS discovery filters and the UE location. For the identified EAS(s) if EAS profile indicates the list of associated devices required in order to serve the UE, then EES checks whether the UE has required associated devices with it or not based on AC profile received in EAS discovery request. If the UE has the required associated devices then EES adds the identified EAS(s) into the list of discovered EAS(s).
When the bundle EAS information is provided, then
- If bundle EAS information includes EAS bundle identifier, the EES identifies all or part of the EAS(s) associated with the same EAS bundle identifier.
- If bundle EAS information includes a list of EASIDs, the EES identifies the EASs which are all or part of the EAS bundle.
If the EEC indicates that service continuity support is required, when identifying the EAS, the EES shall take the indication which ACR scenarios are supported by the AC, the EEC, the EES and the EAS and which of these are preferred by the AC into consideration. The EES may select one EAS and determine whether to perform application traffic influence for this EAS in advance based on AC's service KPI or EAS’s service KPI in desired response time, when the EAS does not perform traffic influence in advance. If the EES determines to perform application traffic influence for this EAS in advance, then the EES will applies the AF traffic influence of the EAS in the 3GPP Core Network before sending EAS discovery response.
If the Prediction expiration time is provided then the EES may determine whether to identify the instantiable but not instantiated EAS as T-EAS based on Prediction expiration time and the predicted EAS deployment time information obtained from ADAES as specified in clause 8.11 of TS 23.436 [28] or from local configured maximum EAS deployment time. The EES determines remaining EAS instantiation time or EAS instantiation completion time based on the timing receiving EAS instantiation is in progress from ECSP management system and the predicted EAS deployment time. Furthermore, if EES received the indication which the EAS instantiation is in progress, then the EES determines whether to identify the EAS which instantiation in progress as T-EAS based on Prediction expiration time and remaining EAS instantiation time information.
When EAS discovery filters are not provided, then:
- if available, the EES identifies the EAS(s) based on the UE-specific service information at the EES and the UE location;
- EES identifies the EAS(s) by applying the ECSP policy (e.g. based only on the UE location),
When EAS discovery filters contain Application group profile, the EES checks whether information about common EAS and related Application Group ID is available or not. If the common EAS information related to the Application Group ID is:
- not available at the EES, then based on the policy if EES needs to select the common EAS, the EES identifies an EAS for the Application Group ID based on the provided EAS discovery filters such as KPIs, UE-specific service information or the ECSP policy. Furthermore, the EES stores the common EAS information and related Application Group ID. The EES may also subscribe and get the notifications of candidate DNAIs for the UEs of the group as described in 3GPP TS 23.548 [20], 3GPP TS 23.501 [2], the EES identifies the EAS(s) taking the candidate DNAIs into account, e.g., an EAS where its DNAI in the EAS profile is common to all UEs in the group.
- available at the EES, then the EES provides information of that EAS as result for EAS discovery, or the EES identifies the EAS based on the provided EAS discovery filters and the UE location when E2E response time is received, furthermore:
- when the UE in the overlapping area between the EDNs of common EAS and EAS registered to this EES (e.g. UE can connect either common EAS or EAS registered to this EES which is UE is in the common EAS service area), the EES identifies either common EAS or EAS registered to this EES for the UE based on the UE location, EAS discovery filters, application group profile, common EAS KPI as specified in table 8.2.5 and EDN information, registered EAS KPI as specified in table 8.2.5 and EDN information, and inter-EAS communication performance of these EASs and checks whether the common EAS or EAS registered to this EES can provide better E2E response time.
- When the UE is not in the overlapping area between the EDNs of common EAS and EAS registered to this EES (e.g. UE is not in the common EAS service area), then the EES identifies candidate EAS(s) for the group from its registered EASs based on UE location, EAS discovery filters and application group profile and checks if the candidate EAS(s) can satisfy the E2E response time considering the response time of the candidate EAS(s), the response time of available common EASs in other EDNs and the inter-EAS communication latency. If the E2E response time cannot be satisfied, the EES will respond the EEC in step 3 indicating such a rejection reason; otherwise, the EES selects a candidate EAS and interacts with the ECS-ER to store it as common EAS as described in clause 8.20.2.3.
NOTE 3: The EES can use ADAE analytics service in 3GPP TS 23.436 [28] or local configuration to obtain application performance (e.g. latency) for inter-EAS communication.
NOTE 4: If the current EDN has overlapping area (where UE resides) with other EDN and both EDNs has available common EAS which can satisfy E2E response time, which EAS to select can take number of UEs in the group connected to available common EAS into consideration.
NOTE 5: The EES may have previously determined and stored the common EAS for Application group ID, or the EES may have received the common EAS selection information for Application group ID during the common EAS announcement procedure.
When the ECS-ER is not available and the EES selects the common EAS, the selected common EAS shall be announced to other EES(s) as per procedure specified in clause 8.19.
When the ECS-ER is available, if the common EAS information related to the Application Group ID is:
- not available at the ECS-ER, the EES identifies one EAS for the group and interacts with the ECS-ER to store the common EAS information as described in clause 8.20.2.3. The EES may also subscribe and get the notifications of candidate DNAIs for the UEs of the group as described in 3GPP TS 23.548 [20], 3GPP TS 23.501 [2], the EES identifies the EAS(s) taking the candidate DNAIs into account, e.g., an EAS where DNAI in the EAS profile is common to all UEs in the group.
- available at the ECS-ER, then the ECS-ER returns the common EAS information to the EES as described in clause 8.20.2.3, or the EES identifies the EAS based on the provided EAS discovery filters and the UE location when E2E response time is received, furthermore:
- When the UE is in the overlapping area between the EDNs of common EAS and EAS registered to this EES (e.g. UE can connect either common EAS or EAS registered to this EES, which is UE is in the common EAS service area), the EES identifies either common EAS or EAS registered to this EES for the UE based on the UE location, EAS discovery filters, application group profile, common EAS KPI as specified in table 8.2.5 and EDN information, registered EAS KPI as specified in table 8.2.5 and EDN information, and inter-EAS communication performance of these EASs and checks whether the common EAS or EAS registered to this EES can provide better E2E response time.
- When the UE is not in the overlapping area between the EDNs of common EAS and EAS registered to this EES (e.g. UE is not in the common EAS service area), the EES checks with the ECS-ER about all available common EASs and their corresponding EDNs by using procedure defined in clause 8.20.2.2. If no common EAS is available for the group, EES identifies one EAS for the group and interacts with the ECS-ER to store the common EAS information as described in clause 8.20.2.3. If current EDN already has available common EAS, the EES selects such common EAS considering UE location, EAS discovery filters, application group profile and the required E2E response time. If there are available common EASs in other EDNs, the EES identifies candidate EAS(s) for the group from its registered EASs and checks if the candidate EAS(s) can satisfy the E2E response time considering UE location, EAS discovery filters, application group profile, and the response time of the candidate EAS(s), the response time of available common EASs in other EDNs and the inter-EAS communication latency. If the E2E response time cannot be satisfied, the EES will respond the EEC in step 3 indicating such a rejection reason; otherwise, the EES selects a candidate EAS and interacts with the ECS-ER to store it as common EAS as described in clause 8.20.2.3.
NOTE 6: The EES can use ADAE analytics service in 3GPP TS 23.436 [28] or local configuration to obtain application performance (e.g. latency) for inter-EAS communication.
NOTE 7: Details of the UE-specific service information and how it is available at the EES is out of scope.
NOTE 8: Both steps are evaluated prior to sending a response.
Upon receiving the request from the EEC, the EES may also collect edge load analytics from ADAES (as specified in clause 8.8.2 of TS 23.436 [28]) or performance data from OAM to find whether the EAS(s) satisfies the Expected AC service KPIs or the Minimum required AC Service KPIs.
Upon receiving the request from the EEC, if the EEC does not indicate EAS Instantiation Triggering Suppress in the EAS Discovery request, the EES may trigger the ECSP management system to instantiate the EAS that matches with EAS discovery filter IEs (e.g. ACID) as in clause 8.12.
Otherwise, upon receiving the request from the EEC, if the EEC indicates EAS Instantiation Triggering Suppress in the EAS Discovery request and the EES supports such capability, the EES determines not triggering the ECSP management system to instantiate the EAS and may determine Instantiable EAS Information for EAS(s) that are instantiable but not yet instantiated and match the EAS discovery filter IEs. Instantiable EAS Information is provided in the EAS Discovery response and includes the EASID(s) and, for each EASID, the status indicating whether the EAS is instantiated or instantiable but not yet instantiated.
If the EEC provides in the EAS discovery request the EAS selection request indicator, the EES selects EAS satisfying the EAS discovery filter or based on other information (e.g. ECSP policy) as described above (if no EAS discovery filter received), and then provides the selected EAS information to the EEC in the discovered EAS list of EAS discovery response.
NOTE 9: Without EAS selection request indication, the EES handling is as per R17 procedure.
If the tunnel information is received, the EES additionally takes it into consideration in identifying EAS(s). If no tunnel information is received, the EES additionally takes the N6 tunnel (e.g. L2TP) information from 3GPP core network (via NEF user plane path management service as described in 3GPP TS 23.501 [2], clause 5.6.7) or NATed UE IP address (e.g. by local UPF based on local configuration) and EAS IP address information into consideration in identifying EAS(s). For instance, the IP address(es) of identified EAS(s) needs to be topologically close to the IP address of the tunnel server, local UPF (based on NATed UE IP address) or NATed UE for optimal N6 route.
3. If the processing of the request was successful, the EES sends an EAS discovery response to the EEC, which includes information about the discovered EASs and Instantiable EAS Information. For discovered EASs, this includes endpoint information. If the EES perform traffic influence for EAS(s) in step 2, then the discovered EAS(s) is with optimized traffic route. Depending on the EAS discovery filters received in the EAS discovery request, the response may include additional information regarding matched capabilities, e.g. service permissions levels, KPIs, AC locations(s) that the EASs can support, ACR scenarios supported by the EAS, etc., and regarding satisfied energy requirements, e.g., supported energy type. The EAS discovery response may contain a list of EASs and Instantiable EAS Information with EAS instantiation completion time. This list may be based on EAS discovery filters containing a Geographical or Topological Service Area, e.g. a route, included in the EAS discovery request by the EEC. When the discovered EAS is for a certain application group, then the Application Group ID is also included in the response message. If the discovered EAS is registered to another EES, then the EES endpoint of the EES where the discovered EAS is registered is also included in the response message.
When the EES accepts the request and determines to trigger the EAS instantiation, then the response may indicate that the EAS instantiation is in progress so that the detailed EAS profile information will be available later. Then the EES determines the remaining EAS instantiation time or EAS instantiation completion time based on the timing receiving EAS instantiation in progress ECSP management system and the predicted EAS deployment time. When EEC receives the EAS instantiation in progress indication, the EEC may send EAS discovery subscription request message if not subscribed yet or send EAS discovery request message later to the EES for obtaining updated EAS information.
If the EES is unable to determine the EAS information using the inputs in the EAS discovery request, UE-specific service information at the EES or the ECSP policy, the EES shall reject the EAS discovery request and respond with an appropriate failure cause.
If the EEC is not registered with the EES, and ECSP policy requires the EEC to perform EEC registration prior to EAS discovery, the EES shall include an appropriate failure cause in the EAS discovery response indicating that EEC registration is required.
If the UE location and predicted/expected UE locations, provided in the EAS discovery request, are outside the Geographical or Topological Service Area of an EAS, then the EES shall not include that EAS in the discovery response. The discovery response may include EAS(s) that cannot serve the UE at its current location if a predicted/expected UE location was provided in the EAS discovery request.
Upon receiving the EAS discovery response, if the EEC selects an EAS which is instantiated (i.e., an EAS profile was provided), the EEC uses the endpoint information for routing of the outgoing application data traffic to EAS(s), as needed, and may provide necessary notifications to the AC(s). The EEC may use the border or overlap between EAS Geographical Service Areas for service continuity purposes. The EEC may cache the EAS information (e.g. EAS endpoint) for subsequent use and avoid the need to repeat step 1. If the Lifetime IE is included in the response, the EEC may cache the EAS information only for the duration specified by the Lifetime IE.
Upon receiving the EAS discovery response, if the EEC selects an EAS which is instantiable but not yet instantiated (i.e. an EAS profile is not provided), the EEC sends the EAS information provisioning request indicating the selected EASID as in clause 8.15. If the EAS discovery response message contains the EAS(s) information along with r EAS instantiation completion time which the EAS instantiation in progress, then the EEC determines whether to identify the EAS which instantiation in progress as T-EAS based on Prediction expiration time and remaining EAS instantiation time/EAS instantiation completion time. If the EEC determines to identify the EAS for which instantiation is in-progress, the EEC may retry to discover the EAS and send the EAS discovery request again after waiting for the time as included in the response message, otherwise, the EAS discovery may fail.
NOTE 10: Within the duration specified by the Lifetime IE, the cached EAS Profile can be updated (e.g. according to notifications from the EES for changes of EAS information due to EAS status change) or the cached EAS Profile can be invalidated due to new EAS information discovery (e.g. due to UE mobility). The EEC can update or invalidate the cached EAS information (e.g. on PDU Session Release or Modification Command).
NOTE 11: The AC can cache the EAS information (e.g. EAS endpoint) for subsequent use. In the case of the cached information needing to be updated or invalidated, the mechanisms for the EEC to notify the AC is up to implementation and is not specified in the current release of the present document.
NOTE 12: The EEC can use the EAS information provided by the discovery procedure to perform service continuity planning, for example when ultra-low latency ACR is required.
If the EAS discovery request fails, the EEC may resend the EAS discovery request, taking into account the received failure cause. If the failure cause indicated that EEC registration is required, the EEC shall perform an EEC registration before resending the EAS discovery request.
NOTE 13: As long as a proper EAS (e.g. considering expected AC service KPIs included in EAS discovery request) is discovered and selected by the EES, EEC of a constraint UE can stop sending EAS discovery to rest candidate EES(s), and provide the selected EAS information to AC.
The information flow for EAS discovery as specified in clause 8.5.3.2 of 3GPP TS 23.558 [aa8] is enhanced as follows (new text in bold italics):
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.2.4 Solution evaluation
| Editor’s Note: This part is for further update.
6.3 Solution #3: Enhance Edge Services for Support Energy Saving
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.3.1 Solution Description
| According to the key issue #3, there is a need for EDGE application enablement layer enhance to support energy saving. The services procedures and information flows introduced in 3GPP TS 23.558 [8] for edge services, can be enhanced to support the energy saving requirements.
The procedure for EAS discovery as specified in clause 8.5.2.2 of 3GPP TS 23.558 [8] is enhanced as follows (new text in bold italics):
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.5.3.2 EAS discovery request
| Table 8.5.3.2-1 describes information elements for the EAS discovery request. Table 8.5.3.2-2 provides further detail about the EAS Discovery Filter information element.
Table 8.5.3.2-1: EAS discovery request
Information element
Status
Description
Requestor identifier
M
The ID of the requestor (e.g. EECID)
UE Identifier
O
The identifier of the UE (i.e., GPSI)
Security credentials
M
Security credentials resulting from a successful authorization for the edge computing service.
EAS discovery filters
O
Set of characteristics to determine required EASs, as detailed in Table 8.5.3.2-2.
UE location
O
The location information of the UE. The UE location is described in clause 7.3.2.
List of UE IDs
O
List of UE IDs in an Application Group, applicable for S-EAS or S-EES triggered EAS discovery request.
Serving MNO information (NOTE 2)
O
The serving MNO information (e.g. MNO name, PLMN ID) which is serving the subscriber.
Target DNAI (NOTE 1)
O
Target DNAI information which can be associated with potential T-EAS(s)
EEC Service Continuity Support
O
Indicates if the EEC supports service continuity or not. The IE also indicates which ACR scenarios are supported by the EEC or, if this message is sent by the EEC to discover a T‑EAS, which ACR scenario(s) are intended to be used for the ACR.
EES Service Continuity Support (NOTE 1)
O
The IE indicates if the S-EES supports service continuity or not. The IE also indicates which ACR scenarios are supported by the S-EES or, if the EAS discovery is used for an S‑EES executed ACR according to clause 8.8.2.5, which ACR scenario is to be used for the ACR.
EAS Service Continuity Support (NOTE 1)
O
The IE indicates if the S-EAS supports service continuity or not. The IE also indicates which ACR scenarios are supported by the S-EAS or, if the EAS discovery is used for an S‑EAS decided ACR according to clause 8.8.2.4, which ACR scenario is to be used for the ACR.
EAS Instantiation Triggering Suppress
O
Indicates to the EES that EAS instantiation triggering should not be performed for the current request, and Instantiable EAS Information (e.g. instantiated, instantiable but not be instantiated yet) is to be provided in response.
EAS selection request indicator
O
Indicates the request for EAS selection support from the EES (e.g., for constrained device).
Indication of service continuity planning
O
Indicates that this EAS discovery request is triggered for service continuity planning.
Prediction expiration time
O
The estimated time the UE may reach the Predicted/Expected UE location or EAS service area at the latest. This IE is used by EES as analytics input to get edge load analytics information from ADAES service as described in clause 8.8 of TS 23.436 [28].
Tunnel information
O
It includes service provider ID, the endpoint address (e.g. IP address) of the tunnel server associated with application(s).
NOTE 1: This IE shall not be included when the request originates from the EEC.
NOTE 2: This IE shall be included if edge node sharing is used.
Table 8.5.3.2-2: EAS discovery filters
Information element
Status
Description
List of AC characteristics (NOTE 1)
O
Describes the ACs for which a matching EAS is needed.
> AC profile (NOTE 2)
M
AC profile containing parameters used to determine matching EAS. AC profiles are further described in Table 8.2.2-1.
> Application group profile (NOTE 6)
O
Application group profile associated with the AC Profile, as defined in Table 8.2.11-1.
List of EAS characteristics (NOTE 1, NOTE 3)
O
Describes the characteristic of required EASs.
> EASID
O
Identifier of the required EAS.
> Application Group ID
O
Application group identifier as defined in 7.2.11.
> EAS content synchronization support
O
Indicates if the EAS content synchronization support is required or not.
> Bundle ID (NOTE 5)
O
A list of EASIDs or a bundle ID as described in clause 7.2.10.
> List of EASIDs (NOTE 5)
O
A list of EASIDs specific to a particular EAS bundle.
> Bundle type (NOTE 4)
O
Type of the EAS bundle as described in clause 7.2.10
> EAS bundle requirements (NOTE 4)
O
Requirements associated with the EAS bundle as described in clause 8.2.10.
> EAS provider identifier
O
Identifier of the required EAS provider
> EAS type
O
The category or type of required EAS (e.g. V2X, UAV, application enabler)
> EAS schedule
O
Required availability schedule of the EAS (e.g. time windows)
> EAS Geographical Service Area (NOTE 6)
O
Location(s) (e.g. geographical area, route) where the EAS service should be available.
> EAS Topological Service Area (NOTE 6)
O
Topological area (e.g. cell ID, TAI) for which the EAS service should be available. See possible formats in Table 8.2.7-1.
> Service continuity support
O
Indicates if the service continuity support is required or not.
> Service permission level
O
Required level of service permissions e.g. trial, gold-class
> Service feature(s)
O
Required service features e.g. single vs. multi-player gaming service
> Energy type
O
Indicates what energy type support is required, e.g. non-renewable energy, renewable energy.
NOTE 1: Either "List of AC characteristics" or "List of EAS characteristics" shall be present.
NOTE 2: "Preferred ECSP list" IE shall not be present.
NOTE 3: The "List of EAS characteristics" IE must include at least one optional IE, if used as an EAS discovery filter.
NOTE 4: When EAS discovery request is sent by the EEC, this IE shall not be included.
NOTE 5: “Bundle ID" and "List of EASIDs" shall not both be present.
NOTE 6: If application group profile IE is present, the expected group geographic service area IE present in the application group profile is preferentially used. If there is no EAS satisfied the expected group geographic service area or EAS does not provide better E2E response time, then the UE location and other service area IEs are considered.
The information flow of EAS Profile for EAS registration and discovery as specified in clause 8.2.4 of 3GPP TS 23.558 [aa8] is enhanced as follows (new text in bold italics):
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.2.4 EAS Profile
| An EAS Profile includes information about an EAS used to describe services and service characteristics offered.
NOTE: Information elements in the EAS Profile are provided by the ASP.
Table 8.2.4-1: EAS Profile
Information element
Status
Description
EASID
M
The identifier of the EAS
EAS Endpoint
M
Endpoint information (e.g. URI, FQDN, IP address) used to communicate with the EAS. This information maybe discovered by EEC and exposed to ACs so that ACs can establish contact with the EAS.
List of EAS bundle information
O
List of EAS bundles to which the EAS belongs and related bundling requirements.
> Bundle ID
(NOTE 3)
O
Bundle ID as described in clause 7.2.10.
> List of EAS IDs
(NOTE 2, NOTE 3)
O
List of the EAS IDs of the EASs to be invoked by the EAS for an EAS driven association of EASs.
> Bundle type
M
Type of the EAS bundle as described in clause 7.2.10
> Main EASID
O
Indicate which EAS in a bundle takes the main EAS service role.
> EAS bundle requirements
O
Requirements associated with the EAS bundle as described in clause 8.2.10.
ACID(s)
O
Identifies the AC(s) that can be served by the EAS
EAS Provider Identifier
O
The identifier of the ASP that provides the EAS.
Allowed MNO information (NOTE 4)
O
Information of the allowed operator (e.g. MNO name, PLMN ID) from which its subscriber can consume the EAS services
EAS Type
O
The category or type of EAS (e.g. V2X, UAV, application enabler)
EAS description
O
Human-readable description of the EAS
EAS Schedule
O
The availability schedule of the EAS (e.g. time windows)
EAS Geographical Service Area
O
The geographical service area that the EAS serves. ACs in UEs that are located outside that area shall not be served.
EAS Topological Service Area
O
The EAS serves UEs that are connected to the Core Network from one of the cells included in this service area. ACs in UEs that are located outside this area shall not be served. See possible formats in Table 8.2.7-1.
EAS Service KPIs
O
Service characteristics provided by EAS, detailed in Table 8.2.5-1
EAS service permission level
O
Level of service permissions e.g. trial, gold-class supported by the EAS
EAS Feature(s)
O
Service features e.g. single vs. multi-player gaming service supported by the EAS
EAS content synchronization support
O
Indicates if the EAS supports content synchronization between EASs.
EAS Service continuity support
O
Indicates if the EAS supports service continuity or not. This IE indicates which ACR scenarios are supported by the EAS, also indicates the EAS ability (e.g. EAS bundle information) of handling bundled EAS coordinate ACR.
EAS Transport layer service continuity support
O
This IE indicates the EAS service continuity support for seamless transport layer (e.g. TCP/TLS/QUIC) relocation
General context holding time duration (NOTE 1)
O
The time duration that the EAS holds the context before the AC connects to the EAS in case of ACR for service continuity planning. It is an indication of the time the EAS holds the application context for a UE to move to its service area after receiving an ACR notification from the EES following an ACR request from the EEC.
List of EAS DNAI(s)
O
DNAI(s) associated with the EAS. This IE is used as Potential Locations of Applications in clause 5.6.7 of 3GPP TS 23.501 [2].
It is a subset of the DNAI(s) associated with the EDN where the EAS resides.
List of N6 Traffic Routing requirements
O
The N6 traffic routing information and/or routing profile ID corresponding to each EAS DNAI.
EAS Availability Reporting Period
O
The availability reporting period (i.e. heartbeat period) that indicates to the EES how often it needs to check the EAS's availability after a successful registration.
EAS Status
O
The status of the EAS (e.g. enabled, disabled, overload warning etc.)
List of associated devices
O
List of associated devices (e.g. haptic device, joy stick) required along with UE in order to provide service to the user. It includes device type.
Energy type
O
Indicates what energy type is supported by the EAS, e.g. non-renewable energy, renewable energy.
NOTE 1: Since the EASID of the EAS identifies the type of the application (e.g. SA6Video, SA6Game etc) as described in clause 7.2.4, "General context holding time duration" determined by EAS can depend on the EASID (type of the application).
NOTE 2: This IE may be provided when only bundle ID is provided, and the bundle type indicates the proxy bundle.
NOTE 3: At least one of the IEs shall be present if EAS bundle information is provided.
NOTE 4: For edge node sharing scenario, in order to restrict the access to the subscriber of the partner operator, this IE should only include MNO information of the leading operator.
NOTE: The EAS Transport layer service continuity support can be used in EAS discovery, e.g. as described in 3GPP TS 23.433 [26] for SEALDD server acting as EAS, which can further support the EAS IP replacement function.
Editor’s Note: How EES gets the information of energy type supported by an EAS is FFS.
6.A3.2 Architecture impacts
Editor’s Note: The architecture impacts of the solution is FFS.
6.A3.3 Solution evaluation
Editor’s Note: The evaluation of the solution is FFS.
6.x4 Solution #x4: Support of AIMLE client selection based on energy consumption
6.x4.1 Solution Description
This solution addresses Key Issue #4 – open issue #3 on how to use energy consumption information for AIMLE client selection. The solution proposes enhancements to the ML model information management procedure, and to procedures involving AIMLE client selection, to enable the storage and use of energy consumption information for AIMLE client selection.
The enhancements enable an AIMLE client to provide energy consumption capabilities during registration. Energy consumption capabilities include the maximum energy budget for performing AIML operations, and the VAL UE’s power profile. The VAL UE power profile indicates the degree of energy consumption of a VAL UE for performing AI/ML operations.
The enhancements also enable an ML model provider to include energy consumption information during ML model storage. The energy consumption information includes expected energy consumption values for performing AIML operations using an ML model. The expected energy consumption values are provided per power profile.
Both the AIMLE energy budget and the ML model energy consumption information are used for AIMLE client selection (e.g., request or subscription). During AIMLE client selection, the AIMLE server receives energy consumption requirements from a requestor. The energy consumption requirements include an indication to consider energy consumption during AIMLE client selection, and the maximum total energy consumption that is acceptable to the requestor. The AIMLE server uses the energy budget and energy consumption information to identify and select the ML models and AIMLE clients that meet the energy consumption requirements of the requestor.
6.x4.1.1 Impact to ML model information management
The ML model information management procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.3.2 Architecture impacts
| Editor’s Note: The architecture impacts of the solution is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.3.3 Solution evaluation
| Editor’s Note: The evaluation of the solution is FFS.
6.4 Solution #4: Support of AIMLE client selection based on energy consumption
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1 Solution Description
| This solution addresses Key Issue #4 – open issue #3 on how to use energy consumption information for AIMLE client selection. The solution proposes enhancements to the ML model information management procedure, and to procedures involving AIMLE client selection, to enable the storage and use of energy consumption information for AIMLE client selection.
The enhancements enable an AIMLE client to provide energy consumption capabilities during registration. Energy consumption capabilities include the maximum energy budget for performing AIML operations, and the VAL UE’s power profile. The VAL UE power profile indicates the degree of energy consumption of a VAL UE for performing AI/ML operations.
The enhancements also enable an ML model provider to include energy consumption information during ML model storage. The energy consumption information includes expected energy consumption values for performing AIML operations using an ML model. The expected energy consumption values are provided per power profile.
Both the AIMLE energy budget and the ML model energy consumption information are used for AIMLE client selection (e.g., request or subscription). During AIMLE client selection, the AIMLE server receives energy consumption requirements from a requestor. The energy consumption requirements include an indication to consider energy consumption during AIMLE client selection, and the maximum total energy consumption that is acceptable to the requestor. The AIMLE server uses the energy budget and energy consumption information to identify and select the ML models and AIMLE clients that meet the energy consumption requirements of the requestor.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.1 Impact to ML model information management
| The ML model information management procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.11.4.1 ML model information storage request
| Table 8.11.4.1-1 describes the information flow from the AIMLE server to the ML repository or from the AIMLE client/VAL server to the AIMLE server as a request for the ML model information storage.
Table 8.11.4.1-1: ML model information storage request
Information element
Status
Description
Requestor Identity
M
The identity of the requestor performing the request.
Security Credentials
O
(NOTE 1)
Security credentials of the requestor performing the request.
ML model
O
(NOTE 2)
The ML model to be stored in the ML repository.
ML model address
O
(NOTE 2)
The address (e.g., a URL or an FQDN) of the ML model file.
ML model information
M
Provides information of the ML model, as described in Table 8.11.4.1-2.
NOTE 1: This information is needed if the requestor is at the VAL service provider domain.
NOTE 2: At least one of these information elements shall be provided.
Table 8.11.4.1-2: ML model information
Information element
Status
Description
ML model identifier
M
An identifier for the ML model
ADAE Analytics ID
O
Represents ADAE analytics ID for which the model can be used.
ML model size
O
Indicates the size of the ML model.
ML model source identifier
O
The identifier of ML model source (e.g., VAL server ID, VAL client ID) that stored the model in the ML repository.
VAL service ID(s)
O
Identify the VAL service ID(s).
Domain
O
Specifies domain for which the model can be used (e.g., for speech recognition, image recognition, video processing, location prediction, etc.).
List of allowed vendors
O
(NOTE 1)
Indicates which vendors that are allowed to use the ML model and thereby also are interoperable to the model.
ML model interoperability information
O
(NOTE 1)
Represents the vendor-specific information that conveys, e.g., requested model file format, model execution environment, input/output parameters of the ML model, etc. The encoding, format, and value of ML Model Interoperable Information is not specified since it is vendor specific information, and is agreed between vendors, if necessary for sharing purposes.
ML Model phase
O
(NOTE 1)
Represents the ML model phase, e.g., in training, trained, re-training, deployed.
> Observed performance
O
(NOTE 2)
Provides information on the performance of the model e.g. accuracy, or application-specific performance metrics (if ML model is in trained or deployment phase).
> Training information
O
(NOTE 2)
If the ML model is in trained or deployed phase: Information on the data that has been used to train the model (e.g. data sources, volume, freshness), and the base model ID in case of Transfer Learning.
> Indication of continuous model training
O
Indicates whether the model can be continuously trained or not.
> Continuous model training parameter
O
Parameters required for continuous model training.
ML model storage and discovery requirements
O
(NOTE 1)
Represents the requirements for the ML repository for the ML model storage and discovery.
> Storage duration
O
Represents the ML model storage duration time. When the storage duration time is expired, the stored ML model and the related information shall be deleted.
> Security and access requirements
O
Represents the information on security requirements for storing the ML model information and the ML model access requirements (e.g., publicly available, private use only, or available for the list of VAL server IDs or VAL client IDs, time period and location access). If the access requirement is private use only, then the model is not discoverable by other consumers.
ML model usage requirements
O
Represents the requirements for using the ML model (e.g. for inference or for training). The requirements are used by the AIMLE server to determine whether an AIMLE client is capable of using the model based on comparing the requirements with information in the AIMLE client profile in Table 8.7.3.1-2.
Energy consumption information
O
Energy consumption information associated with the ML model.
> List of AI/ML operations
M
The AI/ML operations (e.g., ML model training, model transfer, model inference, model offload and model split).
>> Expected energy budget
M
The expected energy budget (e.g., watt-hour, joules) associated with performing an AI/ML operation. The expected energy budget relates energy consumption values to a VAL UE power profile as described in Table 8.7.3.2-2.
NOTE 1: At least one of these information elements shall be provided.
NOTE 2: This IE is included only if trained ML model is available.
6.x4.1.2 Impact to AIMLE client registration
The AIMLE client registration procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.2 Impact to AIMLE client registration
| The AIMLE client registration procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.7.2.2 AIMLE client registration
| Pre-conditions:
1. The AIMLE client has been pre-configured or has discovered the address (e.g., URI) of the AIMLE server.
2. The AIMLE client has been pre-configured with an AIMLE client profile.
Figure 8.7.2.2-1: AIMLE client registration
1. The AIMLE client sends an AIMLE client registration request to the AIMLE server, the registration request includes information as described in Table 8.7.3.2-1. The AIMLE client indicates in the registration request its AI/ML capabilities such as supported ML model types and supported AI/ML operations, supported AIMLE client task capabilities with compute and task performance capabilities, and energy consumption capabilities to assist with performing AIMLE client discovery and AIMLE client selection.
2. The AIMLE server validates the registration request and performs an authentication and authorization check to determine if the AIMLE client is permitted to register to the AIMLE server and participate in AI/ML operations. Upon successful authorization, the AIMLE server saves the context of the AIMLE client registration in the ML repository.
3. The AIMLE server returns an AIMLE client registration response to the AIMLE client with the status of the request.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.7.3.2 AIMLE client registration request
| Table 8.7.3.2-1 shows the request sent by an AIMLE client to an AIMLE server for the AIMLE client registration request.
Table 8.7.3.2-1: AIMLE client registration request
Information element
Status
Description
Requestor identifier
M
The identifier of the requestor.
List of supported profiles
M
Supported AIML client profile(s). For each client profile provided in the list, the supported service information is provided.
> AIMLE client profile
M
Information about the capability of the AIMLE client to support AI/ML operations for the VAL service ID as detailed in Table 8.7.3.2-2.
> List of supported services
M
List of VAL service IDs with corresponding permissions.
> VAL service ID
M
The identifier of the VAL service.
> Service permission level
O
Service permission level (e.g., premium resource usage, standard resource usage, limited resource usage).
Table 8.7.3.2-2: AIMLE client profile
Information element
Status
Description
Supported AI/ML model types
O
AI/ML model types supported by the AIMLE client (e.g., decision trees, linear regression, neural networks).
Supported AI/ML operations
M
AI/ML operations supported by the AIMLE client such as ML model training, model transfer, model inference, model offload, model split, continue perform intermediate AI/ML operation/task).
AIMLE client time schedule configurations
O
Indicates the availability schedule of the AIMLE client for the AIML service, e.g., the AIMLE client is (not) available to participate in the AIML operations (e.g. ML model training) in the given time slot(s) and/or day(s) of the week.
AIMLE client location configurations
O
Indicates the location-based configurations of the AIMLE client for the AIML service, e.g., the AI/ML member is (not) available to participate in the AI/ML operations in the given locations represented by coordinates, civic addresses, network areas, or VAL service area ID.
AIMLE client capabilities
M
AIMLE client capability information (e.g. ML application type, allowed resource usage level).
Dataset availability
O
Dataset availability such as dataset size, age, list of dataset features, and dataset identifiers.
Data capabilities
O
A list of data capabilities such as the type of data that can be collected (e.g. raw data), supported data processing capabilities (e.g. processed data), and supported exploratory data analysis functions.
AIMLE client task capability
O
(NOTE 1)
Indicates the AIML task performing capabilities. It includes compute capabilities (e.g., high, low), task performance preference capabilities (e.g., Green task, energy-efficient, low costs)
Energy consumption capabilities
O
The energy consumption capabilities of the AIMLE client.
> Maximum energy budget
O
(NOTE 2)
The maximum energy budget (e.g., watt-hour, joules) permitted for performing AI/ML operations.
> Power profile
O
(NOTE 2)
The power rating of the VAL UE that represents the VAL UE’s processing capabilities (e.g., entry-level, mid-tier, premium, high performance, high efficiency). The power rating is used to determine the usage rate of energy per unit time that factors into the determination of energy consumption and is static.
NOTE 1: The Green and Energy-efficient task performance may not be applicable to a UE.
NOTE 2: At least one of these information elements shall be present.
6.x4.1.3 Impact to AIMLE client discovery
The AIMLE client selection procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.3 Impact to AIMLE client discovery
| The AIMLE client selection procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.8.3.1 AIMLE client discovery request
| Table 8.8.3.1-1 shows the request sent by a VAL server to an AIMLE server for the AIMLE client discovery procedure.
Table 8.8.3.1-1: AIMLE client discovery request
Information element
Status
Description
Requestor identity
M
The identifier of the requestor (e.g., VAL server).
AIMLE client discovery criteria
M
Discovery criteria for finding suitable AIMLE clients for AI/ML operations as detailed in Table 8.8.3.1-2.
Number of required AIMLE clients
O
Indicates the requested number of AIMLE clients to be discovered based on the discovery criteria.
Table 8.8.3.1-2: AIMLE client discovery criteria
Information element
Status
Description
Service requirement
M
Information about the required service
> VAL Service ID
M
VAL Service ID that the client is required to support. This identifies the service associated with the requester.
> Service permission level
O
Required corresponding service permission level (e.g., premium resource usage standard resource usage, limited resource usage).
Requested ML model types
O
Requested ML model types (decision trees, linear regression, neural networks).
Requested AIML operations
M
Requested role for AI/ML operations such as model training, model transfer, model inference, model offload, model split.
Application layer AIMLE client capabilities
M
Application layer AIMLE client capability information (e.g., ML application type like FL/TL/SL, client availability to support AIML operations at the UE, AIMLE drop off rate).
Dataset requirements
O
Information about dataset.
> Dataset availability
O
Dataset availability, including dataset identifiers, dataset size, age, list of dataset features.
> Data capabilities
O
A list of data capabilities such as the type of data that can be collected (e.g., raw data), supported data processing capabilities (e.g., processed data) and supported exploratory data analysis functions.
AIML client task capability requirements
O
Indicates the AIML task requirements to discover the AIML clients for performing AIML tasks.
It includes compute capabilities (e.g., high, low), task performance preference capabilities (e.g. Green task, Energy-efficient, low costs)
AIMLE client velocity
O
Indicates the AIMLE client velocity. It includes mobile (e.g., high, low), static.
Location information
O
Indicates the location information (e.g., Cell Identity, Tracking Area Identity, GPS Coordinates or civic addresses, VAL service area ID) to discover the AIMLE clients.
AIMLE client QoS requirements
O
Indicates the AIMLE client QoS information (like PLR, bandwidth, latency jitter) with the corresponding threshold(s) and threshold matching direction(s) (e.g., above or below) to discover the AIMLE clients.
Energy consumption requirements
O
(NOTE 1)
The energy consumption requirements of the requestor.
> Maximum energy budget
M
The maximum energy budget (e.g., watt-hour, joules) that is acceptable to the requestor.
NOTE 1: This information element applies to AIMLE client selection.
6.x4.1.4 Impact to AIMLE client selection
The AIMLE client selection procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.4 Impact to AIMLE client selection
| The AIMLE client selection procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.9.2.1 AIMLE client selection
| Pre-conditions:
1. AIMLE clients that support AI/ML operations have registered with the AIMLE server and included their AIMLE client profiles and optionally a list of supported services.
2. The AIMLE server can access a ML repository to obtain AIMLE client profiles and supported services associated with the AIMLE clients.
Figure 8.9.2.1-1: AIMLE client selection
1. A VAL server sends an AIMLE client selection request to an AIMLE server to select AIMLE clients available for participation in AI/ML operations (e.g., is available and have required data to train an ML model). The AIMLE client selection request includes information as described in Table 8.9.3.1-1.
2. The AIMLE server performs authentication and authorization checks to determine if the requestor is able to select AIMLE clients.
3. If the requestor is authorized, the AIMLE server performs AIMLE client selection based on the provided inputs as described below.
For VAL server selection, the AIMLE server receives a list of AIMLE client IDs in the request and selects the AIMLE clients in the list as candidate AIMLE clients.
For AIMLE server selection, the AIMLE server retrieves a list of clients with AI/ML capabilities from the ML repository. From the list of clients, the AIMLE server selects a list of candidate AIMLE clients, whose client profiles fulfil the AIMLE client selection criteria. The AIMLE server may use SEAL (LM service), NEF (e.g. MonitoringEvent API), and NWDAF (e.g. UE mobility analytics as defined in clause 6.7.2 of 3GPP TS 23.288 [2]) capabilities to assist the AIMLE client selection.
The AIMLE server determines UEs by using their identifiers to determine their location and then selects only those that fulfil the location requirements specified in the AIMLE client selection criteria. The AIMLE server may use SEAL-LMS (as in 3GPP TS 23.434 [5] clause 9.3.9, or 3GPP TS 23.434 [5] clause 9.3.10) or 3GPP 5G Core Network Services (such as GMLC as in 3GPP TS 23.273 [13] and NEF as in 3GPP TS 23.273 [13] or 3GPP TS 23.502 [9]) to determine the UEs which fulfil the location requirement.
The AIMLE server may then determine the QoS parameters for the AIML traffic session between the requestor and the candidate AIMLE client(s) and configure the AI/ML traffic session(s) via SEALDD (Sdd_RegularTransmission API) or NEF services (AfSessionWithQoS API). The AIMLE server determines the application QoS parameters based on the VAL Service ID.
NOTE 1: To determine the application QoS parameters based on the VAL service ID, the AIMLE server can use the VAL service ID associated AIMLE client QoS requirements in the AIMLE client selection criteria provided in step 1.
If the AIMLE client selection criteria include energy consumption requirements, the AIMLE server retrieves a list of ML model information from the ML repository using the ML model information discovery procedure as described in clause 8.11.3. The AIMLE server uses information from the AIMLE client selection criteria as filtering criteria for ML model information discovery.
If energy consumption requirements are included, the AIMLE server selects the candidate AIMLE clients that fulfil the expected energy consumption needs for the retrieved ML models. The expected energy consumption needs are based on the requested AI/ML operations and on the AIMLE client’s power profile.
If the energy consumption requirements include a maximum energy budget, the AIMLE server identifies and selects the ML models and candidate AIMLE clients that can perform the requested AI/ML operations without exceeding the maximum energy budget.
NOTE 2: If the available AIMLE clients (as determined by the AIMLE server) that fulfil discovery criteria are less than the required number of AIMLE clients for AI/ML operation (e.g., split AI/ML), then the AIMLE server can discover the remaining required AIMLE clients from other AIMLE servers over AIML-E reference point and include them in the AIMLE client selection response message.
NOTE 3: The AIMLE server can reuse SEAL group management for any necessary group management.
4. The AIMLE server performs AIMLE client participation procedure with each candidate AIMLE client as described in clause 8.10.
For all candidate AIMLE clients that agreed to participate in AI/ML operations, the AIMLE server selects the AIMLE clients and assigns an AIMLE client set identifier for the selected clients. The AIMLE client set may then be used for training a ML model.
NOTE 4: If a required minimum number of AIMLE clients for AI/ML operation is provided in the AIMLE client selection request and the number of AIMLE clients that agreed to participate in AI/ML operations is less than this number, an AIMLE client set identifier will not be assigned and the status will be set to fail in the AIMLE client selection response in step 5.
5. The AIMLE server sends an AIMLE client selection response that includes information in Table 8.9.3.2-1.
6.x4.1.5 Impact to AIMLE client selection subscription
The AIMLE client selection subscription procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.5 Impact to AIMLE client selection subscription
| The AIMLE client selection subscription procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.13.2.2 AIMLE client selection subscription and notification
|
Figure 8.13.2.2-1: AIMLE client selection subscription and notification
1. A VAL server sends an AIMLE client selection subscription request to the AIMLE Server. The AIMLE client selection subscription request includes information as described in Table 8.13.3-1 which includes selection criteria.
2. The AIMLE server validates the AIMLE client selection subscription request. The AIMLE server further performs authentication and authorization checks to determine if the requestor is able to subscribe to the selected AIMLE client selection subscription request.
3. The AIMLE server sends the AIMLE client selection subscription response to the VAL server.
4. The AIMLE server monitors AIMLE clients whether they fulfil the selection criteria as provided in step 1. The AIMLE server interacts with the NEF and/or SEAL services (including SEALDD) to establish monitoring. The AIMLE server utilizes SEAL-LMS (as in 3GPP TS 23.434 [5] clause 9.3.11, or 3GPP TS 23.434 [5] clause 9.3.12) or 3GPP 5G Core Network Services (such as GMLC as in 3GPP TS 23.273 [13] and NEF as in 3GPP TS 23.273 [13] or 3GPP TS 23.502 [9]) to establish monitoring of UEs entering or present in the target location provided in the location information in the selection criteria.
5. The AIMLE Server obtains the identifiers of the AIMLE clients from the monitoring and selects the clients that fulfil the selection criteria and remove the AIMLE clients which do not fulfil the selection criteria. The AIMLE server uses the location monitoring for selecting UEs that fulfil the location criteria and removing UEs which cease to fulfil the location criteria as provided in the location information in the AIMLE client selection criteria. The AIMLE Server may determine the application QoS parameters (e.g. bandwidth, latency, jitter) for the AIML traffic session between the VAL server and the selected AIMLE client and configure the AIML traffic session(s) via SEALDD (Sdd_RegularTransmission API) or NEF services (AfSessionWithQoS API). When the AIMLE clients no more meet the criteria, the QoS adjustment is reversed.
The AIMLE server may determine the application QoS parameters based on the VAL Service ID.
If the AIMLE client selection criteria include energy consumption requirements, the AIMLE server retrieves a list of ML model information from the ML repository using the ML model information discovery procedure as described in clause 8.11.3. The AIMLE server uses information from the AIMLE client selection criteria as filtering criteria for ML model information discovery.
If energy consumption requirements are included, the AIMLE server selects the candidate AIMLE clients that can fulfil the expected energy consumption needs of the ML model. The expected energy consumption needs are based on the requested AI/ML operations and on the AIMLE client’s power profile.
If the energy consumption requirements include a maximum energy budget, the AIMLE server identifies and selects the ML models and candidate AIMLE clients that can perform the requested AI/ML operations without exceeding the maximum energy budget.
If a desired service in the selection criteria is ceased to be provided by the client or its profile change so that it no longer meets the selection criteria, the AIMLE server removes the AIMLE clients which ceases to fulfil the criteria and selects other clients that fulfil selection criteria.
NOTE 1: To determine the application QoS parameters based on the VAL service ID, the AIMLE server can use the VAL service ID associated AIMLE client QoS requirements in the AIMLE client selection criteria provided in step 1.
NOTE 2: In case of AIMLE server determining the application QoS parameters based on the VAL service ID, how it does this is up to implementation.
6. The AIMLE Server notifies the VAL server about the selected and re-selected AIMLE clients e.g., the AIMLE Client A is re-selected and replaced by AIMLE Client B.
6.x4.1.6 Impact to AIMLE client participation
The AIMLE client participation procedure and information flows in 3GPP TS 23.482 [r2348210] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.1.6 Impact to AIMLE client participation
| The AIMLE client participation procedure and information flows in 3GPP TS 23.482 [10] are enhanced (highlighted in bold italics) as follows.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.10.3.1 AIMLE client participation request
| Table 8.10.3.1-1 shows the request sent by the AIMLE server to each AIMLE client selected for AIMLE client participation procedure.
Table 8.10.3.1-1: AIMLE client participation request
Information element
Status
Description
Requestor identity
M
The identifier of the requestor.
AIMLE client set identifier
M
An identifier for the AIMLE client set.
Add/remove indicator
M
Indicator for adding/removing the AIMLE client to/from the AIMLE client set.
ML model ID
M
The ML model ID to use for the AI/ML operation.
AI/ML operations
M
A list of AI/ML operations (e.g., training) required to be performed.
Expected energy budget
O
The expected energy budget (e.g., watt-hour, joules) associated with performing the AI/ML operations.
Operational schedule
O
Schedule for the AI/ML operations.
Dataset requirement
M
Dataset requirements for the AI/ML operations. Requirements includes dataset identifier, dataset size and age, and/or dataset features.
Service requirement
M
Information about the required Service including its VAL service ID and its service permission level (e.g., premium resource usage standard resource usage, limited resource usage)
6.x4.2 Architecture impacts
This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
6.x4.3 Solution evaluation
This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.x5 Solution #x5: AIMLE Enhancement of Considering Renewable Energy
6.x5.1 Solution Description
The proposed solution solve the problem listed in KI#1 and KI#4, and proposes to consider renewable energy in AIMLE.
As indicated in 3GPP TS 28.310 [4], the renewable energy and renewable energy factor have the following definitions:
Renewable energy: energy from renewable non-fossil sources.
NOTE 15: This definition is taken from 3GPP TS 22.261 [31].
NOTE 16: Examples of renewable energy sources include wind, solar, aerothermal, geothermal, hydrothermal and ocean energy, hydropower, biomass, landfill gas, sewage treatment plant gas and biogases
NOTE 17: A renewable energy source is an energy source not depleted by extraction as it is naturally replenished at a rate faster than it is extracted ISO/IEC 30134-3:2016 [33].
NOTE 18: Criteria to categorize an energy as renewable can differ among jurisdictions, based on local environmental or other reasons ISO/IEC 30134-3:2016 [33].
Renewable energy factor: ratio of the renewable energy to the total energy.
NOTE 19: This definition is taken from ISO/IEC 30134-3:2016 [33].
Energy can be classified into two major types: renewable energy and non-renewable energy. Renewable energy sources include solar energy, hydropower, wind energy, biomass energy, wave energy, tidal energy, ocean thermal energy, geothermal energy, etc. They can be recycled and regenerated in nature. It is an inexhaustible and endless energy source that regenerates automatically without human intervention.
In order to save the energy consumption and reduce the carbon emission, the operators may consider using renewable energy or green energy to replace traditional energy. So, in the current AIMLE procedure, the following enhancements can be added:
For AIMLE related procedure:
- The FL member capabilities can include whether renewable energy support or not. It is more preferred to select the FL members that support renewable energy to perform such as AI model training. This capability can be indicated during the registration procedure.
- Also, the capability that support renewable energy can be set as the Member selection criteria. During the ML training procedure, the member that supports the renewable energy can have the priority selection.
6.x5.2 Architecture impacts
In this solution, it doesn’t change the overall architecture of Application enablement architecture of AIMLE in section 5 of 3GPP TS 23.482 [x10]. But the following enhancements are added:
- The FL member can register itself of the new capability that support renewable energy supply.
- The subscriber can be notified with that new FL member is added to the ML repository with the energy supply of renewable energy.
- And during member selection procedure, the Member selection criteria can be enhanced that whether renewable energy support or not.
Editor’s nNote: How FL members know whether renewable energy is supported or not is FFS.
6.x5.3 Procedure
6.x5.3.1 Procedure of FL member registration procedure
This procedure is the same as the procedure in section 8.4.2 of 3GPP TS 23.482 [x10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.2 Architecture impacts
| This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.4.3 Solution evaluation
| This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.5 Solution #5: AIMLE Enhancement of Considering Renewable Energy
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.1 Solution Description
| The proposed solution solve the problem listed in KI#1 and KI#4, and proposes to consider renewable energy in AIMLE.
As indicated in 3GPP TS 28.310 [4], the renewable energy and renewable energy factor have the following definitions:
Renewable energy: energy from renewable non-fossil sources.
NOTE 15: This definition is taken from 3GPP TS 22.261 [31].
NOTE 16: Examples of renewable energy sources include wind, solar, aerothermal, geothermal, hydrothermal and ocean energy, hydropower, biomass, landfill gas, sewage treatment plant gas and biogases
NOTE 17: A renewable energy source is an energy source not depleted by extraction as it is naturally replenished at a rate faster than it is extracted ISO/IEC 30134-3:2016 [33].
NOTE 18: Criteria to categorize an energy as renewable can differ among jurisdictions, based on local environmental or other reasons ISO/IEC 30134-3:2016 [33].
Renewable energy factor: ratio of the renewable energy to the total energy.
NOTE 19: This definition is taken from ISO/IEC 30134-3:2016 [33].
Energy can be classified into two major types: renewable energy and non-renewable energy. Renewable energy sources include solar energy, hydropower, wind energy, biomass energy, wave energy, tidal energy, ocean thermal energy, geothermal energy, etc. They can be recycled and regenerated in nature. It is an inexhaustible and endless energy source that regenerates automatically without human intervention.
In order to save the energy consumption and reduce the carbon emission, the operators may consider using renewable energy or green energy to replace traditional energy. So, in the current AIMLE procedure, the following enhancements can be added:
For AIMLE related procedure:
- The FL member capabilities can include whether renewable energy support or not. It is more preferred to select the FL members that support renewable energy to perform such as AI model training. This capability can be indicated during the registration procedure.
- Also, the capability that support renewable energy can be set as the Member selection criteria. During the ML training procedure, the member that supports the renewable energy can have the priority selection.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.2 Architecture impacts
| In this solution, it doesn’t change the overall architecture of Application enablement architecture of AIMLE in section 5 of 3GPP TS 23.482 [10]. But the following enhancements are added:
- The FL member can register itself of the new capability that support renewable energy supply.
- The subscriber can be notified with that new FL member is added to the ML repository with the energy supply of renewable energy.
- And during member selection procedure, the Member selection criteria can be enhanced that whether renewable energy support or not.
Editor’s Note: How FL members know whether renewable energy is supported or not is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.3 Procedure
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.3.1 Procedure of FL member registration procedure
| This procedure is the same as the procedure in section 8.4.2 of 3GPP TS 23.482 [10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.4.2 Procedure on FL member registration
| Figure 8.4.2-1 illustrates the procedure where the registration of a candidate FL member happens via the ML repository, serving as AIML service registry.
Figure 8.4.2-1: Procedure for registration on FL member registry
1. The candidate FL member (e.g., VAL server via AIMLE server or AIMLE server) sends an FL member registration request to the ML repository for registering to the ML repository which acts as the AIML service registry. The request may include capability whether this FL member supports the renewable energy as energy supply or not.
Editor’s Nnote: How does FL member knows the capability of renewable energy supporting is FFS.
2. The ML repository validates the received request and generates the identity and other security related information for all the FL members listed in the registration request.
3. The ML repository sends the generated information in the FL member registration response message to the candidate FL member.
The procedure for AIMLE client to register to ML repository as FL member is introduced in clause 8.7.2.2.
6.x5.3.2 Procedure of FL-related event subscription of renewable energy
This procedure is the same as the procedure in section 8.5.2 of 3GPP TS 23.482 [x10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.3.2 Procedure of FL-related event subscription of renewable energy
| This procedure is the same as the procedure in section 8.5.2 of 3GPP TS 23.482 [10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.5.2 Procedure on subscription for FL related events
| This procedure, as illustrated in Figure 8.5.2-1, describes the subscription for events related to FL member availability.
Pre-conditions:
1. The AIMLE server has the authorization to subscribe for the FL-related events (events described in clause 8.5.4).
Figure 8.5.2-1: Procedure for FL-related event subscription
1. The subscriber (AIMLE server or VAL server (via AIMLE server)) sends an FL-related event subscription request to the ML repository to receive notification of FL related events and in particular the availability of FL member for a target area and time. The subscription includes indicator whether the FL member supports the capability of renewable energy supporting or not.
2. Upon receiving the event subscription request from the AIMLE server, the ML repository checks for the relevant authorization for the event subscription. If the authorization is successful, the ML repository stores the subscription information.
3. The ML repository sends an FL-related event subscription response to the subscriber indicating successful operation.
6.x5.3.3 Procedure of member selection during ML model training considering renewable energy
This procedure is the same as the procedure in section 8.3.2 of 3GPP TS 23.482 [x10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.3.3 Procedure of member selection during ML model training considering renewable energy
| This procedure is the same as the procedure in section 8.3.2 of 3GPP TS 23.482 [10] with the following enhancement:
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.3.2 Procedure for ML model training
| Figure 8.3.2-1 illustrates the procedure for AIMLE server to support ML model training based on the request from the VAL server.
Figure 8.3.2-1: ML model training
1. The VAL server sends an ML model training request to AIMLE server, requesting to assist in its ML model training. This request consists of ML model information or ML model requirement information, capability of renewable energy supporting or not, etc.
2. The AIMLE server checks whether the VAL server is authorized to perform the ML model training request. If no model information is provided but only the model requirement information is provided in step 1, the AIMLE server identifies and selects the appropriate ML model for training based on the ML model requirement information using the procedure as specified in clause 8.11.3.
3. If the VAL server is authorized, AIMLE server returns the success response, otherwise a failure response indication the reason for failure.
4. The AIMLE server notifies the VAL server to update the list FL/ML clients selected or de-selected for the ML model training or to share the training output or any errors during the training process.
6.x5.4 Solution evaluation
Editor’s Nnote: This part is for further update in the future.
6.x6 Solution #x6: Support of ADAE analytics for AI/ML energy consumption
6.x6.1 Solution Description
This solution addresses Key Issue #4 – open issue #1 on how to collect and use energy consumption data. The solution proposes a new ADAE service procedure in 3GPP TS 23.436 [r2343611] to support AI/ML energy consumption analytics.
6.x6.1.1 Procedure
Figure 6.x6.1.1-1 illustrates the procedure for a consumer (e.g., VAL server, AIMLE server) to subscribe for AI/ML energy consumption analytics generated from data collected from VAL UEs while the VAL UEs are performing AI/ML operations.
Pre-conditions:
1. ADAE Client (ADAEC) is connected to the ADAE Server (ADAES).
2. Data producers (e.g., VAL Clients) may be pre-configured with data producer profiles for the data they can provide. ADAES and ADAEC have discovered available data producers and their data producer profiles.
Figure 6.x6.1.1-1: ADAES support for AI/ML energy consumption analytics
1. An analytics consumer (e.g. VAL Server, AIMLE Server) sends an AI/ML energy consumption analytics subscription request to an ADAES. The subscription request includes an Analytics ID “AI/ML energy consumption analytics”, ML model identifier, AIML operation, energy consumption metrics, and other information as defined in Table 6.x6.1.2.1-1.
2. The ADAES validates the subscription request and performs authentication and authorization checks to determine if the consumer is authorized to subscribe for the requested analytics. If authorization is successful, the ADAES stores the subscription request information, assigns a subscription ID, and sends a subscription response indicating successful subscription and includes the subscription ID.
3. The ADAES maps the analytics ID to a list of data collection event identifiers and determines a list of data producer IDs for the data collection event identifiers. The determination of data producer IDs is based on data producer profiles.
4. The ADAES sends an AI/ML energy consumption data collection subscription request to the ADAECs determined in step 3. The subscription request includes information as defined in Table 6.x6.1.2.4-1. Data collection at the VAL UE(s) reuses the EVEX mechanism defined in 3GPP TS 26.531 [r2653112].
5. Each ADAEC configures the data producers (i.e., VAL client or AIMLE client) for collecting data (e.g., duration of AI/ML operation, % resource utilization for AI/ML operation) for AI/ML energy consumption analytics requested in the data collection subscription request.
6. The ADAEC sends an AI/ML energy consumption data collection subscription response to acknowledge the ADAES. The response includes a status and a subscription ID for the data collection.
7. The data producers perform AI/ML operations when triggered (e.g., ML model training) using the indicated ML model and collect data for AI/ML energy consumption analytics based on the configured data collection requirements.
8. At the completion of AI/ML operation, the ADAEC sends an AI/ML energy consumption data collection notification to the ADAES. The data notification includes the data collection subscription ID and the information as defined in Table 6.x6.1.2.6-1.
9. The ADAES performs analytics relevant operations to generate the AI/ML energy consumption analytics based on the data received from the ADAECs. The ADAES uses the data collected by the ADAECs to calculate the AI/ML energy consumption analytics.
NOTE: How the collected data are used to calculate energy consumption is up to implementation.
10. The ADAES sends an AI/ML energy consumption analytics notification to the consumer. The analytics notification includes the analytics subscription ID and the information as defined in Table 6.x6.1.2.3-1.
6.x6.1.2 Information flows
6.x6.1.2.1 AI/ML energy consumption analytics subscription request
Table 6.x6.1.2.1-1 describes the information flow from the consumer (e.g. VAL server, AIMLE server) as a request or update request for AI/ML energy consumption analytics.
Table 6.x6.1.2.1-1: AI/ML energy consumption analytics subscription request
Information element
Status
Description
Requestor ID
M
The identifier of the consumer (e.g., VAL server, AIMLE client).
Analytics ID
M
The identifier of the analytics event. This ID can be for example "AI/ML energy consumption analytics".
Analytics type
M
The type of analytics for the event (e.g., statistics or predictions).
ML model identifier
M
The identifier of the ML model for which analytics are determined.
AIML operation
M
The desired AIML operations (e.g., ML model training, inference) for which analytics are determined.
Energy consumption metrics
M
The formula and necessary metrics for calculating the energy consumption based on factors that impact energy consumption for an AIML operation using a particular ML model.
Target VAL UE IDs
O
The identifier(s) of VAL UE(s) for which the analytics subscription applies.
Target data producer IDs
O
The identifier(s) of VAL clients, AIMLE clients or FL members to target for collecting data for energy consumption analytics.
Target data producer profile criteria
O
Characteristics of the data producers to be used.
Reporting requirements
O
It describes the requirements for analytics reporting. This requirement may include e.g. the type and frequency of reporting (periodic or event triggered), the reporting periodicity in case of periodic, and reporting thresholds.
Area of Interest
O
The geographical or service area for which the subscription request applies.
Preferred confidence level
O
The level of accuracy for the analytics service (in case of prediction).
Time validity
O
The time validity of the subscription request.
6.x6.1.2.2 AI/ML energy consumption analytics subscription response
Table 6.x6.1.2.2-1 describes the information elements for the AI/ML energy consumption analytics subscription response from the ADAES to the consumer.
Table 6.x6.1.2.2-1: AI/ML energy consumption analytics subscription response
Information element
Status
Description
Result
M
The result of the analytics subscription request (positive or negative acknowledgement).
Subscription ID
M
An identifier for the subscription.
6.x6.1.2.3 AI/ML energy consumption analytics notification
Table 6.x6.1.2.3-1 describes the information flow from the ADAES to the consumer (e.g. VAL Server, AIMLE Server) as a notification for the AI/ML energy consumption analytics.
Table 6.x6.1.2.3-1: AI/ML energy consumption analytics notification
Information element
Status
Description
Subscription ID
M
The identifier for the subscription.
Analytics ID
M
The identifier of the analytics event. This ID can be for example " AI/ML energy consumption analytics".
ML model identifier
M
The ML model identifier for which analytics are determined.
Energy consumption information
M
The range of energy consumption values for a particular ML model with a certain AIML operation. The output can be an average, minimum, and/or maximum value for the energy consumption. The output may be a prediction or a statistical value.
List of AIML operations
M
The AIML operations for which the analytics are determined.
> Energy consumption
M
The range of energy consumption values for performing an AI/ML operation.
>> Maximum consumption
O
The maximum energy consumption value.
>> Average consumption
O
The average energy consumption value.
>> Minimum consumption
O
The minimum energy consumption value.
Target analytics period
O
The time validity of the analytics.
Confidence level
O
The level of accuracy for the analytics output (e.g. prediction).
6.x6.1.2.4 AI/ML energy consumption data collection subscription request
Table 6.x6.1.2.4-1 describes information elements for the AI/ML energy consumption data collection subscription request from the ADAES to the ADAEC.
Table 6.x6.1.2.4-1: AI/ML energy consumption data collection subscription request
Information element
Status
Description
Requestor ID
M
The identifier of the consumer.
Data Collection Event ID
M
The identifier of the data collection event.
Data collection requirements
M
The requirements for data collection and for information to include in the data notification.
> ML model identifier
M
The identifier of the ML model for which data is collected.
> AIML operations
M
The desired AI/ML operations (e.g., ML model training, inference) for which data is collected.
> Time period
M
The time period of data collection.
> Location or area information
O
The location or area of the data collection.
6.x6.1.2.5 AI/ML energy consumption data collection subscription response
Table 6.x6.1.2.5-1 describes information elements for the AI/ML energy consumption data collection subscription response from the ADAEC to the ADAES.
Table 6.x6.1.2.5-1: AI/ML energy consumption data collection subscription response
Information element
Status
Description
Result
M
The result of the AI/ML energy consumption data collection subscription request (positive or negative acknowledgement).
Subscription ID
M
An identifier for the subscription.
6.x6.1.2.6 AI/ML energy consumption data notification
Table 6.x6.1.2.6-1 describes information elements for the AI/ML energy consumption data notification from the ADAEC to the ADAES.
Table 6.x6.1.2.6-1: AI/ML energy consumption data collection notification
Information element
Status
Description
Subscription ID
M
The identifier for the subscription.
Data Collection Event ID
M
The identifier of the data collection event.
Data Producer ID
M
The identity of Data Producer.
Collected data
M
The collected data.
> ML model identifier
M
The identifier of the ML model for which data was collected.
> AIML operation
M
The AI/ML operation (e.g., ML model training, inference) for which data was collected.
> Power profile
M
The power rating of the VAL UE that represents the VAL UE’s processing capabilities (e.g., entry-level, mid-tier, premium, high performance, high efficiency). The power rating is used to determine the usage rate of energy per unit time that factors into the determination of energy consumption and is static.
> Elapse time for AI/ML operation
M
The time duration for performing the AI/ML operation.
> % resource utilization for AI/ML operation
M
The percentage utilization of hardware resources (e.g., compute, memory, storage) determined for performing the AI/ML operation.
6.x6.2 Architecture impacts
This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
6.x6.3 Solution evaluation
This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.x7 Solution #x7: ADAES Support for AIMLE client energy sustainability analytics
6.x7.1 Solution Description
This solution introduces a new analytics event for “AIMLE client energy sustainability analytics” as requested from the Consumer who can be the VAL Server or AIMLE server to the ADAES. These analytics may be used for predicting whether an AI/ML client at the VAL UE can be considered a candidate in a ML training or inference task, or FL process based on its expected energy status/capability. Such energy analytics can be used for identifying whether the energy consumption is sustainable for a given AIMLE service, or a recommendation on whether the AIMLE client is applicable or suitable for the AI/ML task based on the energy prediction or sustainability indicator.
This clause describes the procedure for supporting the AIMLE client energy analytics in ADAES as new analytics service.
•
Figure 6.x7.1-1: ADAES support for AIMLE client energy analytics
1. The Consumer (AIMLE or VAL server) sends an AIMLE client energy sustainability analytics request to ADAES to perform analytics on the AIMLE client Energy Consumption who is expected to undertake an AIML task, Event ID= “AIMLE client energy sustainability analytics”, for a given service area and a given time window. This request includes a VAL service ID or AIMLE service ID for which the analytics apply, as well as the consumer identifier and the analytics KPI. Such request can include a predicted or expected route of the VAL UE hosting the AIMLE client, and the type of analytics requested. The analytics type can be requesting whether the energy consumption is sustainable for a given AIMLE service.
Such request can also include an AI/ML task or operation or process (ML training or inference or FL task) for which the prediction/stats apply.
2. The ADAES authorizes the consumer’s request.
3. The ADAES maps the AIMLE client energy analytics event ID to a list of data collection event identifiers, and a list of data producer IDs. Such mapping may be preconfigured by OAM or may be determined by ADAES based on the analytics event type/vertical type and/or data producer profile.
ADAES determines and maps the analytics event to the energy data to acquire and the data producers which can be functionalities in the target VAL UE(s) and/or data stored in A-ADRF.
Optionally, the ADAES (or A-ADRF) keeps per AIMLE client. This energy profile keeps stats or energy rating of the AIMLE client or the data source in the UE (e.g. AI app) which can be used as a metric to identify whether the AIMLE client at the VAL UE is energy demanding source; hence applicable or not to be considered as capable of undertaking an AI/ML task.
NOTE: If Energy Profile is used, this information may either be provided by the VAL UE (or the VAL server) as part of the energy data or can be quantified at the ADAES.
4a. The ADAES requests and receives from the VAL UE(s) of interest energy data using the AIMLE client Energy Data Collection API. Such Energy data collection can include measurements or stats on the energy consumption per AIMLE client.
Editor’s Note: Further details on the energy data collection from the VAL UE is FFS.
4b. In addition, or complementary to AIMLE client energy data collection, the ADAES may also request and receive from the ADAEC of the UE, performance statistics related to the application layer or enabler layer sessions (based on existing ADAE capabilities on VAL session analytics).
5. The ADAES calculates the expected AIMLE client energy consumption based on the received energy/ performance data in step 4, based on the request.
6. The ADAES obtains the corresponding trained ML model based on procedure in 3GPP TS 23.482 [10] clause 8.3.2 and performs analytics to derive the predicted UE energy consumption sustainability at the target area and time horizon.
7. The ADAES sends the analytics output via the AIMLE client energy sustainability analytics API to the consumer (VAL or AIMLE server) as analytics response with the analytics output data (based on the type of analytics in the request). The analytics output may include:
• identification on whether the energy consumption is sustainable for a given VAL or AIMLE service, where this service can be an AI/ML service
• a recommendation on whether the AIMLE client is applicable or suitable for the AI/ML task based on the energy prediction or sustainability indicator.
The ADAES and/or the AIMLE (as consumer) may also store the energy data and/or analytics to a repository being A-ADRF or ML repository (if this applies to an ML task).
Based on step 7, the consumer (AIMLE or VAL server) can use these analytics as input to trigger the pro-active selection or re-selection or removal of the AIMLE client from the list of UEs which are capable of undertaking an AI/ML process/task/operation, given the energy factor.
6.x7.2 Architecture impacts
This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
6.x7.3 Solution evaluation
This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.Y18 Solution #8: Network Slice Energy Optimization based on Energy Saving VAL Server Policy
6.Y18.1 Solution Description
6.Y18.1.0 General
This solution does not add new procedure. Instead, it proposes introducing a new VAL server policy in clause 9.5, 'Network slice optimization based on VAL server policy', 3GPP TS 23.435 [zz13]. The new policies enable NSCE server performs slice modifications based on monitoring energy-related information, such as energy consumption and the renewable energy ratio, at the slice level.
The high-level principles are as follows:
1. The VAL server provisions energy saving policies to the NSCE server.
• Based on the monitored energy consumption of a network slice - categorized by energy source (renewable or non-renewable) - over a specific duration (such as one hour) or as accumulated over a longer period (for example, a calendar month), slice modification is triggered when the measured value exceeds or falls below a defined threshold. This modification may involve reducing the slice's maximum throughput, slightly increasing allowable latency, or decreasing redundancy levels (such as removing a redundant UPF instance or redundant RAN cell).
• Based on the monitored renewable energy ratio of a network slice, a slice modification with predefined parameters is triggered when the energy ratio or crosses a specified threshold. Such modifications may include increasing the slice’s maximum throughput, raising allowable latency, expanding the maximum number of supported UEs, or increasing the maximum number of PDU sessions.
2. The NSCE server monitors a slice's energy usage by gathering energy-related data from the slice's OAM system.
3. Based on the defined triggers, the NSCE server carries out the actions outlined in energy-saving policies, aiming to either reduce the slice’s energy consumption or encourage the use of renewable energy sources.
The following clauses specify network slice energy optimization based on energy saving VAL server policy.
6.Y18.1.1 Procedure
Figure 6.Y18.1.1-1 outlines the procedure, where Steps 1-4 represent existing operations. Step 5 introduces a new function in which the NSCE server monitors energy-related information for a network slice. Step 6 involves the NSCE server taking action to adjust the slice, either to conserve energy or to encourage the use of renewable energy sources.
Figure 6.Y18.1.1-1: VAL server policy provision and enforcment for energy optimization
1. The consumer (VAL, SEAL Server) sends a VAL server policy provisioning request to the NSCE server by following clause 9.5 of 3GPP TS 23.435 [zz13]. New energy saving related policies (see examples in clause 6.Y18.1.2) are added:
a. Based on the monitored energy consumption of a network slice - categorized by energy source (renewable or non-renewable) - over a specific duration (such as one hour) or as accumulated over a longer period (for example, a calendar month), slice modification is triggered when the measured value exceeds or falls below a defined threshold. This modification may involve reducing the slice's maximum throughput, slightly increasing allowable latency, or decreasing redundancy levels (such as removing a redundant UPF instance or redundant RAN cell).
b. Based on the monitored renewable energy ratio of a network slice, a slice modification with predefined parameters is triggered when the energy ratio or crosses a specified threshold. Such modifications may include increasing the slice’s maximum throughput, raising allowable latency, expanding the maximum number of supported UEs, or increasing the maximum number of PDU sessions.
The action of the above policy may include charging policy update when slice SLA is modified to save energy to attract the VAL server to use the policies.
2. The NSCE server performs a VAL server policy check. This is the same as step 2 in clause 9.5.2.1, 3GPP TS 23.435 [zz13].
3. The NSCE server performs Slice Harmonization. This step is the same as clause 9.5.2.1.4 Policy harmonization, 3GPP TS 23.435 [zz13].
4. The NSCE server sends back a VAL server policy provisioning response to the Consumer (VAL/ASP), confirming policy provision. This is the same as step 2 in clause 9.5.2, 3GPP TS 23.435 [zz13].
5. The NSCE server initiates the retrieval of energy consumption data from the slice OAM (3GPP TS 28.554 [6]) to monitor energy-related information, including but not limited to the following parameters:
a. slice accumulated energy consumption refers to summing the average energy usage recorded during each slice OAM measurement interval (for example, every 15 minutes) by the OAM system and aggregating it into the total energy consumption over an extended period, such as a calendar month
b. slice average energy consumption: e.g. Energy-Consumption/hour
c. slice renewable energy ratio: e.g. 30%
Editor’s Note: Whether the NSCE collect the energy data directly or by using existing enabler service (e.g. A-DCCF) is FFS.
6. The NSCE server carries out slice modifications in accordance with the policy action defined in Step 1, through its interaction with the Slice OAM and the 5GC.
6.Y18.1.2 Information Flows
6.Y18.1.2.1 General
The following information flows are specified for VAL server policy.
6.Y18.1.2.2 Policy of Reducing Energy Consumption
Table 6.Y18.1.2.2-1 shows the request sent by a consumer to an NSCE server for VAL server policy provisioning procedure.
Table 6.Y18.1.2.2-1: Policy of reducing energy consumption
Information element
Status
Description
Policy
O
Reduce Energy Consumption
>Area of interest
M
The service area for which the policy profile applies, which can be expressed as a geographical area (e.g. geographical coordinates), or a topological area (e.g. a list of TA).
>Trigger event
M
Threshold information, i.e. slices accumulate energy consumption, slices average energy consumption reached the threshold
>Recommended action
M
Examples:
• Modification of PDU sessions / max number of UEs (step for decrease in %).
• Decrease slice capacity (step for decrease in %).
• Decrease packet delay, and increase packet error rate etc.
• Update the energy efficiency parameter.
• Decrease GBR requirements.
• Apply better charging offering for the slice after adjustment.
Lifetime or number of events
M
Time duration or number of times the policy can take action.
Priority
O
Indicates the priority of the policy.
Scheduling period
O
Indicates the scheduling of policy in terms of time.
>Start time
M
Indicates the scheduled start time.
>End time
M
Indicates the scheduled end time.
Preemption
O
Indicates the pre-empt capability of the policy.
6.Y18.1.2.3 Policy of Promoting renewable energy consumption
Table 6.Y18.1.2.3-1 shows the request sent by a consumer to an NSCE server for VAL server policy provisioning procedure.
Table 6.Y18.1.2.3-1: Policy of Promoting renewable energy consumption
Information element
Status
Description
Policy
O
Promote renewable energy usage
>Area of interest
M
The service area for which the policy profile applies, which can be expressed as a geographical area (e.g. geographical coordinates), or a topological area (e.g. a list of TA).
>Trigger event
M
Threshold information, i.e. when renewable energy reached a threshold %
>Recommended action
M
Examples:
• Modification of PDU sessions/max number of UEs (step for increase in %).
• Increase slice capacity.
Lifetime or number of events
M
Time duration or number of times the policy can take action.
Priority
O
Indicates the priority of the policy.
Scheduling period
O
Indicates the scheduling of policy in terms of time.
>Start time
M
Indicates the scheduled start time.
>End time
M
Indicates the scheduled end time.
Preemption
O
Indicates the pre-empt capability of the policy.
6.Y18.2 Architecture impacts
Editor’s Note: The architecture impacts of the solution is FFS.
6.Y18.3 Solution evaluation
Editor’s Note: The evaluation of the solution is FFS.
6.Y9 Solution #Y9: Enhancements to ADAE DN Energy Efficiency Analytics
6.Y9.1 Solution Description
In 3GPP TS 23.436 [yy11], the DN Energy Efficiency analytics, ADAES provides analytics on the energy consumption /efficiency of an edge platform (including the EESs / EASs). The DN energy analytics is performed per DNN/ DNAI and may be used to trigger the application server migration to different cloud. The analytics are based on NWDAF analytics and UPF/DN measurements on user plane load as well as edge/app side measurements on the energy consumption. Information/analytics on renewable energy used for providing application services is needed to support flexibility adjustment of the execution of workloads with GHG emissions. However, analytics relevant to renewable energy used for providing application services is missing. Enhancements to the DN Energy Efficiency analytics is required.
The DN Energy Efficiency analytics as specified in clause 8.18 of 3GPP TS 23.436 [yy11] is enhanced as follows (new text in bold italics):
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.5.4 Solution evaluation
| Editor’s Note: This part is for further update in the future.
6.6 Solution #6: Support of ADAE analytics for AI/ML energy consumption
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1 Solution Description
| This solution addresses Key Issue #4 – open issue #1 on how to collect and use energy consumption data. The solution proposes a new ADAE service procedure in 3GPP TS 23.436 [11] to support AI/ML energy consumption analytics.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.1 Procedure
| Figure 6.6.1.1-1 illustrates the procedure for a consumer (e.g., VAL server, AIMLE server) to subscribe for AI/ML energy consumption analytics generated from data collected from VAL UEs while the VAL UEs are performing AI/ML operations.
Pre-conditions:
1. ADAE Client (ADAEC) is connected to the ADAE Server (ADAES).
2. Data producers (e.g., VAL Clients) may be pre-configured with data producer profiles for the data they can provide. ADAES and ADAEC have discovered available data producers and their data producer profiles.
Figure 6.6.1.1-1: ADAES support for AI/ML energy consumption analytics
1. An analytics consumer (e.g. VAL Server, AIMLE Server) sends an AI/ML energy consumption analytics subscription request to an ADAES. The subscription request includes an Analytics ID “AI/ML energy consumption analytics”, ML model identifier, AIML operation, energy consumption metrics, and other information as defined in Table 6.6.1.2.1-1.
2. The ADAES validates the subscription request and performs authentication and authorization checks to determine if the consumer is authorized to subscribe for the requested analytics. If authorization is successful, the ADAES stores the subscription request information, assigns a subscription ID, and sends a subscription response indicating successful subscription and includes the subscription ID.
3. The ADAES maps the analytics ID to a list of data collection event identifiers and determines a list of data producer IDs for the data collection event identifiers. The determination of data producer IDs is based on data producer profiles.
4. The ADAES sends an AI/ML energy consumption data collection subscription request to the ADAECs determined in step 3. The subscription request includes information as defined in Table 6.6.1.2.4-1. Data collection at the VAL UE(s) reuses the EVEX mechanism defined in 3GPP TS 26.531 [12].
5. Each ADAEC configures the data producers (i.e., VAL client or AIMLE client) for collecting data (e.g., duration of AI/ML operation, % resource utilization for AI/ML operation) for AI/ML energy consumption analytics requested in the data collection subscription request.
6. The ADAEC sends an AI/ML energy consumption data collection subscription response to acknowledge the ADAES. The response includes a status and a subscription ID for the data collection.
7. The data producers perform AI/ML operations when triggered (e.g., ML model training) using the indicated ML model and collect data for AI/ML energy consumption analytics based on the configured data collection requirements.
8. At the completion of AI/ML operation, the ADAEC sends an AI/ML energy consumption data collection notification to the ADAES. The data notification includes the data collection subscription ID and the information as defined in Table 6.6.1.2.6-1.
9. The ADAES performs analytics relevant operations to generate the AI/ML energy consumption analytics based on the data received from the ADAECs. The ADAES uses the data collected by the ADAECs to calculate the AI/ML energy consumption analytics.
NOTE: How the collected data are used to calculate energy consumption is up to implementation.
10. The ADAES sends an AI/ML energy consumption analytics notification to the consumer. The analytics notification includes the analytics subscription ID and the information as defined in Table 6.6.1.2.3-1.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2 Information flows
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.1 AI/ML energy consumption analytics subscription request
| Table 6.6.1.2.1-1 describes the information flow from the consumer (e.g. VAL server, AIMLE server) as a request or update request for AI/ML energy consumption analytics.
Table 6.6.1.2.1-1: AI/ML energy consumption analytics subscription request
Information element
Status
Description
Requestor ID
M
The identifier of the consumer (e.g., VAL server, AIMLE client).
Analytics ID
M
The identifier of the analytics event. This ID can be for example "AI/ML energy consumption analytics".
Analytics type
M
The type of analytics for the event (e.g., statistics or predictions).
ML model identifier
M
The identifier of the ML model for which analytics are determined.
AIML operation
M
The desired AIML operations (e.g., ML model training, inference) for which analytics are determined.
Energy consumption metrics
M
The formula and necessary metrics for calculating the energy consumption based on factors that impact energy consumption for an AIML operation using a particular ML model.
Target VAL UE IDs
O
The identifier(s) of VAL UE(s) for which the analytics subscription applies.
Target data producer IDs
O
The identifier(s) of VAL clients, AIMLE clients or FL members to target for collecting data for energy consumption analytics.
Target data producer profile criteria
O
Characteristics of the data producers to be used.
Reporting requirements
O
It describes the requirements for analytics reporting. This requirement may include e.g. the type and frequency of reporting (periodic or event triggered), the reporting periodicity in case of periodic, and reporting thresholds.
Area of Interest
O
The geographical or service area for which the subscription request applies.
Preferred confidence level
O
The level of accuracy for the analytics service (in case of prediction).
Time validity
O
The time validity of the subscription request.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.2 AI/ML energy consumption analytics subscription response
| Table 6.6.1.2.2-1 describes the information elements for the AI/ML energy consumption analytics subscription response from the ADAES to the consumer.
Table 6.6.1.2.2-1: AI/ML energy consumption analytics subscription response
Information element
Status
Description
Result
M
The result of the analytics subscription request (positive or negative acknowledgement).
Subscription ID
M
An identifier for the subscription.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.3 AI/ML energy consumption analytics notification
| Table 6.6.1.2.3-1 describes the information flow from the ADAES to the consumer (e.g. VAL Server, AIMLE Server) as a notification for the AI/ML energy consumption analytics.
Table 6.6.1.2.3-1: AI/ML energy consumption analytics notification
Information element
Status
Description
Subscription ID
M
The identifier for the subscription.
Analytics ID
M
The identifier of the analytics event. This ID can be for example " AI/ML energy consumption analytics".
ML model identifier
M
The ML model identifier for which analytics are determined.
Energy consumption information
M
The range of energy consumption values for a particular ML model with a certain AIML operation. The output can be an average, minimum, and/or maximum value for the energy consumption. The output may be a prediction or a statistical value.
List of AIML operations
M
The AIML operations for which the analytics are determined.
> Energy consumption
M
The range of energy consumption values for performing an AI/ML operation.
>> Maximum consumption
O
The maximum energy consumption value.
>> Average consumption
O
The average energy consumption value.
>> Minimum consumption
O
The minimum energy consumption value.
Target analytics period
O
The time validity of the analytics.
Confidence level
O
The level of accuracy for the analytics output (e.g. prediction).
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.4 AI/ML energy consumption data collection subscription request
| Table 6.6.1.2.4-1 describes information elements for the AI/ML energy consumption data collection subscription request from the ADAES to the ADAEC.
Table 6.6.1.2.4-1: AI/ML energy consumption data collection subscription request
Information element
Status
Description
Requestor ID
M
The identifier of the consumer.
Data Collection Event ID
M
The identifier of the data collection event.
Data collection requirements
M
The requirements for data collection and for information to include in the data notification.
> ML model identifier
M
The identifier of the ML model for which data is collected.
> AIML operations
M
The desired AI/ML operations (e.g., ML model training, inference) for which data is collected.
> Time period
M
The time period of data collection.
> Location or area information
O
The location or area of the data collection.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.5 AI/ML energy consumption data collection subscription response
| Table 6.6.1.2.5-1 describes information elements for the AI/ML energy consumption data collection subscription response from the ADAEC to the ADAES.
Table 6.6.1.2.5-1: AI/ML energy consumption data collection subscription response
Information element
Status
Description
Result
M
The result of the AI/ML energy consumption data collection subscription request (positive or negative acknowledgement).
Subscription ID
M
An identifier for the subscription.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.1.2.6 AI/ML energy consumption data notification
| Table 6.6.1.2.6-1 describes information elements for the AI/ML energy consumption data notification from the ADAEC to the ADAES.
Table 6.6.1.2.6-1: AI/ML energy consumption data collection notification
Information element
Status
Description
Subscription ID
M
The identifier for the subscription.
Data Collection Event ID
M
The identifier of the data collection event.
Data Producer ID
M
The identity of Data Producer.
Collected data
M
The collected data.
> ML model identifier
M
The identifier of the ML model for which data was collected.
> AIML operation
M
The AI/ML operation (e.g., ML model training, inference) for which data was collected.
> Power profile
M
The power rating of the VAL UE that represents the VAL UE’s processing capabilities (e.g., entry-level, mid-tier, premium, high performance, high efficiency). The power rating is used to determine the usage rate of energy per unit time that factors into the determination of energy consumption and is static.
> Elapse time for AI/ML operation
M
The time duration for performing the AI/ML operation.
> % resource utilization for AI/ML operation
M
The percentage utilization of hardware resources (e.g., compute, memory, storage) determined for performing the AI/ML operation.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.2 Architecture impacts
| This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.6.3 Solution evaluation
| This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.7 Solution #7: ADAES Support for AIMLE client energy sustainability analytics
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.7.1 Solution Description
| This solution introduces a new analytics event for “AIMLE client energy sustainability analytics” as requested from the Consumer who can be the VAL Server or AIMLE server to the ADAES. These analytics may be used for predicting whether an AI/ML client at the VAL UE can be considered a candidate in a ML training or inference task, or FL process based on its expected energy status/capability. Such energy analytics can be used for identifying whether the energy consumption is sustainable for a given AIMLE service, or a recommendation on whether the AIMLE client is applicable or suitable for the AI/ML task based on the energy prediction or sustainability indicator.
This clause describes the procedure for supporting the AIMLE client energy analytics in ADAES as new analytics service.
•
Figure 6.7.1-1: ADAES support for AIMLE client energy analytics
1. The Consumer (AIMLE or VAL server) sends an AIMLE client energy sustainability analytics request to ADAES to perform analytics on the AIMLE client Energy Consumption who is expected to undertake an AIML task, Event ID= “AIMLE client energy sustainability analytics”, for a given service area and a given time window. This request includes a VAL service ID or AIMLE service ID for which the analytics apply, as well as the consumer identifier and the analytics KPI. Such request can include a predicted or expected route of the VAL UE hosting the AIMLE client, and the type of analytics requested. The analytics type can be requesting whether the energy consumption is sustainable for a given AIMLE service.
Such request can also include an AI/ML task or operation or process (ML training or inference or FL task) for which the prediction/stats apply.
2. The ADAES authorizes the consumer’s request.
3. The ADAES maps the AIMLE client energy analytics event ID to a list of data collection event identifiers, and a list of data producer IDs. Such mapping may be preconfigured by OAM or may be determined by ADAES based on the analytics event type/vertical type and/or data producer profile.
ADAES determines and maps the analytics event to the energy data to acquire and the data producers which can be functionalities in the target VAL UE(s) and/or data stored in A-ADRF.
Optionally, the ADAES (or A-ADRF) keeps per AIMLE client. This energy profile keeps stats or energy rating of the AIMLE client or the data source in the UE (e.g. AI app) which can be used as a metric to identify whether the AIMLE client at the VAL UE is energy demanding source; hence applicable or not to be considered as capable of undertaking an AI/ML task.
NOTE: If Energy Profile is used, this information may either be provided by the VAL UE (or the VAL server) as part of the energy data or can be quantified at the ADAES.
4a. The ADAES requests and receives from the VAL UE(s) of interest energy data using the AIMLE client Energy Data Collection API. Such Energy data collection can include measurements or stats on the energy consumption per AIMLE client.
Editor’s Note: Further details on the energy data collection from the VAL UE is FFS.
4b. In addition, or complementary to AIMLE client energy data collection, the ADAES may also request and receive from the ADAEC of the UE, performance statistics related to the application layer or enabler layer sessions (based on existing ADAE capabilities on VAL session analytics).
5. The ADAES calculates the expected AIMLE client energy consumption based on the received energy/ performance data in step 4, based on the request.
6. The ADAES obtains the corresponding trained ML model based on procedure in 3GPP TS 23.482 [10] clause 8.3.2 and performs analytics to derive the predicted UE energy consumption sustainability at the target area and time horizon.
7. The ADAES sends the analytics output via the AIMLE client energy sustainability analytics API to the consumer (VAL or AIMLE server) as analytics response with the analytics output data (based on the type of analytics in the request). The analytics output may include:
• identification on whether the energy consumption is sustainable for a given VAL or AIMLE service, where this service can be an AI/ML service
• a recommendation on whether the AIMLE client is applicable or suitable for the AI/ML task based on the energy prediction or sustainability indicator.
The ADAES and/or the AIMLE (as consumer) may also store the energy data and/or analytics to a repository being A-ADRF or ML repository (if this applies to an ML task).
Based on step 7, the consumer (AIMLE or VAL server) can use these analytics as input to trigger the pro-active selection or re-selection or removal of the AIMLE client from the list of UEs which are capable of undertaking an AI/ML process/task/operation, given the energy factor.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.7.2 Architecture impacts
| This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.7.3 Solution evaluation
| This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.8 Solution #8: Network Slice Energy Optimization based on Energy Saving VAL Server Policy
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1 Solution Description
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.0 General
| This solution does not add new procedure. Instead, it proposes introducing a new VAL server policy in clause 9.5, 'Network slice optimization based on VAL server policy', 3GPP TS 23.435 [13]. The new policies enable NSCE server performs slice modifications based on monitoring energy-related information, such as energy consumption and the renewable energy ratio, at the slice level.
The high-level principles are as follows:
1. The VAL server provisions energy saving policies to the NSCE server.
• Based on the monitored energy consumption of a network slice - categorized by energy source (renewable or non-renewable) - over a specific duration (such as one hour) or as accumulated over a longer period (for example, a calendar month), slice modification is triggered when the measured value exceeds or falls below a defined threshold. This modification may involve reducing the slice's maximum throughput, slightly increasing allowable latency, or decreasing redundancy levels (such as removing a redundant UPF instance or redundant RAN cell).
• Based on the monitored renewable energy ratio of a network slice, a slice modification with predefined parameters is triggered when the energy ratio or crosses a specified threshold. Such modifications may include increasing the slice’s maximum throughput, raising allowable latency, expanding the maximum number of supported UEs, or increasing the maximum number of PDU sessions.
2. The NSCE server monitors a slice's energy usage by gathering energy-related data from the slice's OAM system.
3. Based on the defined triggers, the NSCE server carries out the actions outlined in energy-saving policies, aiming to either reduce the slice’s energy consumption or encourage the use of renewable energy sources.
The following clauses specify network slice energy optimization based on energy saving VAL server policy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.1 Procedure
| Figure 6.8.1.1-1 outlines the procedure, where Steps 1-4 represent existing operations. Step 5 introduces a new function in which the NSCE server monitors energy-related information for a network slice. Step 6 involves the NSCE server taking action to adjust the slice, either to conserve energy or to encourage the use of renewable energy sources.
Figure 6.8.1.1-1: VAL server policy provision and enforcment for energy optimization
1. The consumer (VAL, SEAL Server) sends a VAL server policy provisioning request to the NSCE server by following clause 9.5 of 3GPP TS 23.435 [13]. New energy saving related policies (see examples in clause 6.8.1.2) are added:
a. Based on the monitored energy consumption of a network slice - categorized by energy source (renewable or non-renewable) - over a specific duration (such as one hour) or as accumulated over a longer period (for example, a calendar month), slice modification is triggered when the measured value exceeds or falls below a defined threshold. This modification may involve reducing the slice's maximum throughput, slightly increasing allowable latency, or decreasing redundancy levels (such as removing a redundant UPF instance or redundant RAN cell).
b. Based on the monitored renewable energy ratio of a network slice, a slice modification with predefined parameters is triggered when the energy ratio or crosses a specified threshold. Such modifications may include increasing the slice’s maximum throughput, raising allowable latency, expanding the maximum number of supported UEs, or increasing the maximum number of PDU sessions.
The action of the above policy may include charging policy update when slice SLA is modified to save energy to attract the VAL server to use the policies.
2. The NSCE server performs a VAL server policy check. This is the same as step 2 in clause 9.5.2.1, 3GPP TS 23.435 [13].
3. The NSCE server performs Slice Harmonization. This step is the same as clause 9.5.2.1.4 Policy harmonization, 3GPP TS 23.435 [13].
4. The NSCE server sends back a VAL server policy provisioning response to the Consumer (VAL/ASP), confirming policy provision. This is the same as step 2 in clause 9.5.2, 3GPP TS 23.435 [13].
5. The NSCE server initiates the retrieval of energy consumption data from the slice OAM (3GPP TS 28.554 [6]) to monitor energy-related information, including but not limited to the following parameters:
a. slice accumulated energy consumption refers to summing the average energy usage recorded during each slice OAM measurement interval (for example, every 15 minutes) by the OAM system and aggregating it into the total energy consumption over an extended period, such as a calendar month
b. slice average energy consumption: e.g. Energy-Consumption/hour
c. slice renewable energy ratio: e.g. 30%
Editor’s Note: Whether the NSCE collect the energy data directly or by using existing enabler service (e.g. A-DCCF) is FFS.
6. The NSCE server carries out slice modifications in accordance with the policy action defined in Step 1, through its interaction with the Slice OAM and the 5GC.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.2 Information Flows
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.2.1 General
| The following information flows are specified for VAL server policy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.2.2 Policy of Reducing Energy Consumption
| Table 6.8.1.2.2-1 shows the request sent by a consumer to an NSCE server for VAL server policy provisioning procedure.
Table 6.8.1.2.2-1: Policy of reducing energy consumption
Information element
Status
Description
Policy
O
Reduce Energy Consumption
>Area of interest
M
The service area for which the policy profile applies, which can be expressed as a geographical area (e.g. geographical coordinates), or a topological area (e.g. a list of TA).
>Trigger event
M
Threshold information, i.e. slices accumulate energy consumption, slices average energy consumption reached the threshold
>Recommended action
M
Examples:
• Modification of PDU sessions / max number of UEs (step for decrease in %).
• Decrease slice capacity (step for decrease in %).
• Decrease packet delay, and increase packet error rate etc.
• Update the energy efficiency parameter.
• Decrease GBR requirements.
• Apply better charging offering for the slice after adjustment.
Lifetime or number of events
M
Time duration or number of times the policy can take action.
Priority
O
Indicates the priority of the policy.
Scheduling period
O
Indicates the scheduling of policy in terms of time.
>Start time
M
Indicates the scheduled start time.
>End time
M
Indicates the scheduled end time.
Preemption
O
Indicates the pre-empt capability of the policy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.1.2.3 Policy of Promoting renewable energy consumption
| Table 6.8.1.2.3-1 shows the request sent by a consumer to an NSCE server for VAL server policy provisioning procedure.
Table 6.8.1.2.3-1: Policy of Promoting renewable energy consumption
Information element
Status
Description
Policy
O
Promote renewable energy usage
>Area of interest
M
The service area for which the policy profile applies, which can be expressed as a geographical area (e.g. geographical coordinates), or a topological area (e.g. a list of TA).
>Trigger event
M
Threshold information, i.e. when renewable energy reached a threshold %
>Recommended action
M
Examples:
• Modification of PDU sessions/max number of UEs (step for increase in %).
• Increase slice capacity.
Lifetime or number of events
M
Time duration or number of times the policy can take action.
Priority
O
Indicates the priority of the policy.
Scheduling period
O
Indicates the scheduling of policy in terms of time.
>Start time
M
Indicates the scheduled start time.
>End time
M
Indicates the scheduled end time.
Preemption
O
Indicates the pre-empt capability of the policy.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.2 Architecture impacts
| Editor’s Note: The architecture impacts of the solution is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.8.3 Solution evaluation
| Editor’s Note: The evaluation of the solution is FFS.
6.9 Solution #9: Enhancements to ADAE DN Energy Efficiency Analytics
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.9.1 Solution Description
| In 3GPP TS 23.436 [11], the DN Energy Efficiency analytics, ADAES provides analytics on the energy consumption /efficiency of an edge platform (including the EESs / EASs). The DN energy analytics is performed per DNN/ DNAI and may be used to trigger the application server migration to different cloud. The analytics are based on NWDAF analytics and UPF/DN measurements on user plane load as well as edge/app side measurements on the energy consumption. Information/analytics on renewable energy used for providing application services is needed to support flexibility adjustment of the execution of workloads with GHG emissions. However, analytics relevant to renewable energy used for providing application services is missing. Enhancements to the DN Energy Efficiency analytics is required.
The DN Energy Efficiency analytics as specified in clause 8.18 of 3GPP TS 23.436 [11] is enhanced as follows (new text in bold italics):
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18 Procedure for supporting DN Energy Efficiency analytics
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.1 General
| This clause describes the procedure for DN energy consumption/efficiency analytics, where the analytics are performed based on data collected from one or more DNs and A-ADRF.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.2 Procedure
| Figure 8.18.2-1 illustrates the procedure for DN energy efficiency analytics enablement solution.
Pre-conditions:
1. Data producers (e.g. A-ADRF, EAS, EES) may be pre-configured with data producer profiles (as in Table 8.2.4.8-1) for the data they can provide. ADAES has discovered available data producers and their data producer profiles.
Figure 8.8.2.1-1: ADAES support for DN energy analytics
1. The VAL server sends a DN energy analytics request/subscription request to ADAES to perform analytics on the DN Energy Consumption/Efficiency for one or more DNs/EDNs, Event ID= “DN energy analytics”, “Energy analytics per application service”, or “Ratio of renewable energy analytics”, for a given DN service area (or subarea) and a given time window. The message in the request/subscription request is as described in Table 8.18.3.2-1.
2. The ADAES authorizes the VAL request. The ADAES responses to the analytics request. The message in the response is as described in Table 8.18.3.4-1.
3. The ADAES requests and receives from the EAS /VAL servers hosted at the serving and target DNs (within the VAL service area).
3a. (If Event ID= “DN energy analytics” in step 1) expected application service load and traffic schedules for the ongoing or future sessions within the area. Such data include traffic schedule report for the VAL Server, and this step re-uses the step 3 to 10 of clause 8.8.2.1.
3b. (If Event ID= “Energy analytics per application service”, or “Ratio of renewable energy analytics” in step 1) energy usage data at the DN per application, and/or in a time period, and/or the type of energy resources used.
4. The ADAES calculates the expected energy consumption or efficiency based on the received traffic and load data for the given DNN/DNAI based on the request.
NOTE: How the collected data are used to calculate energy efficiency metric is up to implementation.
5. (Optional) The ADAES obtains the corresponding trained ML model based on procedure in 3GPP TS 23.482 clause 8.3.2 and performs analytics to derive the predicted energy consumption at the target area and time horizon. The analytics outputs can be the statistics or predicted energy consumption / efficiency for the given DNN/DNAI.
6. The ADAES sends a DN energy analytics response/notifications with the energy consumption/efficiency analytics output data to the VAL server. The message in the analytics response/notifications is as described in Table 8.18.3.3-1.
Based on 6, the VAL server can use these analytics as input to trigger pro-actively:
- an application server migration to a different edge cloud or to a centralized cloud as a way of reducing the energy consumption for the edge (if consumption is expected to be very high (e.g. higher than a pre-configured threshold)).
- an application server offboarding and the instantiation of a new server at the target edge/centralized cloud to minimize energy consumption of the edge platform (taking into account the system wide energy efficiency).
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.3 Information flows
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.3.1 General
| The following information flows are specified for DN energy analytics based on clause 8.18.2.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.3.2 DN energy analytics request/subscription request
| Table 8.18.3.2-1 describes information elements for the DN energy analytics request/subscription request from the VAL server to the ADAE server.
Table 8.18.3.2-1: DN energy analytics request/subscription request
Information element
Status
Description
Analytics Consumer ID
M
The identifier of the analytics consumer (VAL server, EAS).
Analytics ID
M
The identifier of the analytics event. This ID can be for example “DN energy analytics”, “Energy analytics per application service”, “Ratio of renewable energy analytics”.
Analytics type
M
The type of analytics for the event, e.g. statistics or predictions.
VAL service ID
O
The identifier of the VAL service for which the analytics is requested.
This IE shall be provided when the Analytics ID is “Energy analytics per application service”.
DNN/DNAI
M
DNN or DNAIs information for which the subscription applies.
Energy Efficiency/Consumption metrics
O
The formula and necessary metrics for calculating the EE based on load and traffic information per DN.
Target data producer profile criteria
O
Characteristics of the data producers to be used.
Renewable energy type
O
Indicate the type of renewable energy for which the analytics applies (e.g., solar, wind).
This IE may be provided when the Analytics ID is “Ratio of renewable energy analytics”. If omit, it indicates all types of renewable energy.
Preferred confidence level
O
The level of accuracy for the analytics service (in case of prediction).
Area of Interest
O
The geographical or service area for which the subscription request applies.
Time period of Interest
O
The time period for which the analytics applies.
This IE shall be provided when the Analytics ID is “Ratio of renewable energy analytics”.
Time validity
O
The time validity of the subscription request.
Reporting requirements
O
It describes the requirements for the energy analytics reporting. This requirement may include e.g. the type and frequency of reporting (periodic or event triggered), the reporting periodicity in case of periodic, the event in case of event triggered (e.g. the energy consumption/efficiency crosses a threshold, the ratio of renewable energy crosses a threshold), and reporting thresholds.
Notification endpoints ID or address
O
The identifier or address of the notification endpoints (e.g. VAL server, VAL client, or other network entities).
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.3.3 DN energy analytics response/notification
| Table 8.18.3.3-1 describes information elements for the DN energy analytics response/notification request from the VAL server to the ADAE server to the VAL server (or to the notification endpoints provided in the request/subscription request).
Table 8.18.3.3-1: DN energy analytics response/notification
Information element
Status
Description
Result
M
The result of the analytics request (positive or negative acknowledgement).
Analytics ID
O
The identifier of the analytics event.
Analytics Output
O
The predictive or statistical parameter, which can be stats or prediction related to the energy consumption or efficiency for the edge platform for a given area/time and based on the event type.
> DNN
M
Identifies the data network name for which analytics information is provided.
> DNAI
M
Identifier of a user plane access to one or more DN(s) of the DN.
> Energy metrics
O
The predicted energy metrics.
>> Energy Consumption (NOTE)
O
The predicted energy consumption per DNAI based on network and edge resource usage
>> Energy Efficiency (NOTE)
O
The energy efficiency per DNAI based on network and edge resource usage (given a certain optimal energy consumption metric, which can be pre-configured).
>> DN Data Volume (NOTE)
O
The predicted data volume per DNAI.
>> Energy per Application service (NOTE)
O
The statistics or predicted energy consumption or energy efficiency per application service based on network and edge resource usage.
>>> VAL service ID
M
The identifier of the VAL service for which the analytics applied.
>> Ratio of renewable energy (NOTE)
O
The statistics or predicted ratio of renewable energy analytics per DNAI or per application service based on network and edge resource usage.
>>> Renewable energy type
O
Indicate the type of renewable energy used (e.g., solar, wind).
If omit, it indicates all types of renewable energy.
>>>> Renewable energy ratio
O
Ratio of the renewable energy.
> Area of Interest
O
The area (topological or geographical or edge area) where the analytics apply.
> Applicable time period
O
The time period that the analytics applies to.
>Confidence level
O
For predictive analytics, the achieved confidence level can be provided.
NOTE: At least one of these shall be present.
Editor’s Note: How to obtain the renewable energy information is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.18.3.4 Response to DN energy analytics request
| Table 8.18.3.4-1 describes information elements for the ADAES responses to the analytics request to the consumer (e.g. VAL server).
Table 8.18.3.4-1: Response to DN energy analytics request
Information element
Status
Description
Successful response (NOTE)
O
(NOTE)
Indicates that the request was successful.
> Subscription Identifier
O
The unique identifier for the event subscription. This IE shall be provided when it’s subscription request.
Failure response (NOTE)
O
(NOTE)
Indicates that the request failed.
> Cause
O
Indicates the cause of request failure (e.g. the energy usage data is not available).
NOTE: One of these IEs shall be present in the message.
6.Y9.2 Architecture impacts
Editor’s Note: The architecture impacts of the solution is FFS.
6.Y9.3 Solution evaluation
Editor’s Note: The evaluation of the solution is FFS.
6.x10 Solution #x10: Support of energy saving for location services
6.x10.1 Solution description
This solution addresses the KI#6.
The following are the procedures to introduce how LMS subscribes the energy related information analysis from ADAES (clause 6.x10.1.1) and how LMS utilizes the obtained energy related information analysis to reduce the energy consumption for the location services (clause 6.x10.1.2).
6.x10.1.1 Procedure of LMS subscription for the energy related information analysis
Figure 6.x10.1.1-1 illustrates the high-level procedure of the LMS subscription for the energy related information analysis.
The LMS may subscribe to the ADAES for the analysis of energy related information.
Figure 6.x10.1.1-1: Procedure of LMS subscription for the energy related information analysis
1. LMS may subscribe the energy information analysis related to location services from ADAES, including the analysis ID (e.g., energy consumption information), the service ID, the target area, the energy type (e.g., energy consumption), the energy information granularity (e.g., per VAL UE per location service), the filtering information (e.g., maximum energy consumption per VAL UE), the target UE ID, the trigger conditions, the time duration, etc.
2. The ADAES check if the LMS is authorized to request the energy information analysis. If the request is authorized, the ADAES responds to the LMS.
3. The ADAES sends the energy information analysis response to the LMS.
4. To obtain the analysis for the energy information, the ADAES may subscribe the energy related information (e.g., energy consumption information) from 3GPP Core network as described in clause 5.51.2.4 of 3GPP TS 23.501 [235013], with the parameters received in service request in step 1.
5. When the trigger condition is met, the 3GPP Core network provides the energy related information (e.g., energy consumption information) to the ADAES per requested granularity (e.g., per VAL UE per location service).
6. Upon receiving the energy related information in step 5, the ADAES utilizes and generates the analysis for energy related information per VAL UE per location service, including the energy information statistic (e.g., maximum/average/minimum energy consumption information over a certain period of time, etc.), historical energy information, energy information predication (e.g., the time point of peak energy consumption), etc.
Editor’s nNote: How the ADAES generates the analysis for energy related information and which information is FFS.
7. The ADAES notifies the energy information analysis to the LMS as service requested, including the analysis results generated in step 6.
6.x10.1.2 Procedure of LMS adjusting the location reporting configuration to save the energy
Figure 6.x10.1.2-1 illustrates the high-level procedure of LMS adjusting the location reporting configuration to save the energy consumption.
Figure 6.x10.1.2-1: Procedure of LMS adjusting the location reporting configuration to save the energy
1. Similar with step 1 of clause 9.3.5 in 3GPP TS 23.434 [2343414], the VAL server sends a location reporting trigger to the LMS to activate a location reporting procedure for obtaining the location information of LMC with the parameters defined in clause 9.3.2.4 of 3GPP TS 23.434 [2343414]. The location reporting trigger is periodical and the request also includes the energy saving indicator which means the LMS should consider saving the energy when performing the location services for the target UE.
2. Upon receiving the service request, the LMS decides to subscribe the energy information analysis from the ADAES for the target UE, following the procedure as described in clause 6.x10.1.1.
The ADAES replies the energy related information analysis for the target UE to the LMS, and may include the energy information statistic (e.g., maximum/average/minimum energy consumption information over a certain period of time), energy information predication (e.g., the time point of peak energy consumption), etc.
3. Based on the energy related information analysis received from ADAES and the service request from VAL server, the LMS determines the UE location reporting configuration. For example, the LMS may decrease the reporting frequency when the UE’s energy consumption almost reaches its daily maximum, and vice versa. Or the LMS may pend the location reporting when the time point of peak energy consumption is coming based on the UE's energy information predication.
4. The LMS may interact with the VAL server regarding the determined UE location reporting configuration and sends the adaptive location reporting configuration provisioning messages as defined in clause 9.3.20 of 3GPP TS 23.434 [2343414].
5. If the VAL server approves the suggested location reporting configuration, the LMS sends such configurations to the LMC as defined in clause 9.3.3 of 3GPP TS 23.434 [2343414]. If the VAL server rejects the suggested location configuration, the LMS may discard the configuration.
6. Considering the received location reporting configuration, the LMC reports the location information to the LMS dynamically when the location reporting trigger is met.
7. The LMS reports the received location information for the target UE to the VAL server.
6.x10.2 Architecture Impacts
This solution has no impact on the existing architecture.
6.x10.3 Solution evaluation
This clause provides an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.x Solution #x: <Title>
Provide a suitable title for the solution.
6.x.1 Solution Description
This clause will provide the description of the solution. Each solution should clearly describe which of the key issues it covers and how.
6.x.2 Architecture impacts
This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
6.x.3 Solution evaluation
This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.9.2 Architecture impacts
| Editor’s Note: The architecture impacts of the solution is FFS.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.9.3 Solution evaluation
| Editor’s Note: The evaluation of the solution is FFS.
6.10 Solution #10: Support of energy saving for location services
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.10.1 Solution description
| This solution addresses the KI#6.
The following are the procedures to introduce how LMS subscribes the energy related information analysis from ADAES (clause 6.10.1.1) and how LMS utilizes the obtained energy related information analysis to reduce the energy consumption for the location services (clause 6.10.1.2).
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.10.1.1 Procedure of LMS subscription for the energy related information analysis
| Figure 6.10.1.1-1 illustrates the high-level procedure of the LMS subscription for the energy related information analysis.
The LMS may subscribe to the ADAES for the analysis of energy related information.
Figure 6.10.1.1-1: Procedure of LMS subscription for the energy related information analysis
1. LMS may subscribe the energy information analysis related to location services from ADAES, including the analysis ID (e.g., energy consumption information), the service ID, the target area, the energy type (e.g., energy consumption), the energy information granularity (e.g., per VAL UE per location service), the filtering information (e.g., maximum energy consumption per VAL UE), the target UE ID, the trigger conditions, the time duration, etc.
2. The ADAES check if the LMS is authorized to request the energy information analysis. If the request is authorized, the ADAES responds to the LMS.
3. The ADAES sends the energy information analysis response to the LMS.
4. To obtain the analysis for the energy information, the ADAES may subscribe the energy related information (e.g., energy consumption information) from 3GPP Core network as described in clause 5.51.2.4 of 3GPP TS 23.501 [3], with the parameters received in service request in step 1.
5. When the trigger condition is met, the 3GPP Core network provides the energy related information (e.g., energy consumption information) to the ADAES per requested granularity (e.g., per VAL UE per location service).
6. Upon receiving the energy related information in step 5, the ADAES utilizes and generates the analysis for energy related information per VAL UE per location service, including the energy information statistic (e.g., maximum/average/minimum energy consumption information over a certain period of time, etc.), historical energy information, energy information predication (e.g., the time point of peak energy consumption), etc.
Editor’s Note: How the ADAES generates the analysis for energy related information and which information is FFS.
7. The ADAES notifies the energy information analysis to the LMS as service requested, including the analysis results generated in step 6.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.10.1.2 Procedure of LMS adjusting the location reporting configuration to save the energy
| Figure 6.10.1.2-1 illustrates the high-level procedure of LMS adjusting the location reporting configuration to save the energy consumption.
Figure 6.10.1.2-1: Procedure of LMS adjusting the location reporting configuration to save the energy
1. Similar with step 1 of clause 9.3.5 in 3GPP TS 23.434 [14], the VAL server sends a location reporting trigger to the LMS to activate a location reporting procedure for obtaining the location information of LMC with the parameters defined in clause 9.3.2.4 of 3GPP TS 23.434 [14]. The location reporting trigger is periodical and the request also includes the energy saving indicator which means the LMS should consider saving the energy when performing the location services for the target UE.
2. Upon receiving the service request, the LMS decides to subscribe the energy information analysis from the ADAES for the target UE, following the procedure as described in clause 6.10.1.1.
The ADAES replies the energy related information analysis for the target UE to the LMS, and may include the energy information statistic (e.g., maximum/average/minimum energy consumption information over a certain period of time), energy information predication (e.g., the time point of peak energy consumption), etc.
3. Based on the energy related information analysis received from ADAES and the service request from VAL server, the LMS determines the UE location reporting configuration. For example, the LMS may decrease the reporting frequency when the UE’s energy consumption almost reaches its daily maximum, and vice versa. Or the LMS may pend the location reporting when the time point of peak energy consumption is coming based on the UE's energy information predication.
4. The LMS may interact with the VAL server regarding the determined UE location reporting configuration and sends the adaptive location reporting configuration provisioning messages as defined in clause 9.3.20 of 3GPP TS 23.434 [14].
5. If the VAL server approves the suggested location reporting configuration, the LMS sends such configurations to the LMC as defined in clause 9.3.3 of 3GPP TS 23.434 [14]. If the VAL server rejects the suggested location configuration, the LMS may discard the configuration.
6. Considering the received location reporting configuration, the LMC reports the location information to the LMS dynamically when the location reporting trigger is met.
7. The LMS reports the received location information for the target UE to the VAL server.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.10.2 Architecture Impacts
| This solution has no impact on the existing architecture.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 6.10.3 Solution evaluation
| This clause provides an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
6.x Solution #x: <Title>
Provide a suitable title for the solution.
6.x.1 Solution Description
This clause will provide the description of the solution. Each solution should clearly describe which of the key issues it covers and how.
6.x.2 Architecture impacts
This clause will provide the architecture impacts of the solution and possible new SA6 capabilities and interfaces.
6.x.3 Solution evaluation
This clause will provide an evaluation of the solution. The evaluation should include the descriptions of the impacts to existing architectures.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 7 Overall Evaluation
| This clause will provide evaluation of different solutions.
|
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8 Conclusions
| |
eb1343f2cdb3d270d65fac7ddb0d0638 | 23.700-44 | 8.1 General conclusions
| This clause will provide general conclusions for the study.
8.2 Conclusions of key issue #x
This clause will provide conclusions for the specific key issue.
Annex A (informative):
Change history
Change history
Date
Meeting
TDoc
CR
Rev
Cat
Subject/Comment
New version
2025-08
SA6#68
TR skeleton as approved by SA6 in S6-253267.
0.0.0
2025-08
SA6#68
Implementation of the following pCRs approved by SA6:
S6-253641, S6-253642, S6-253698, S6-253715, S6-253716, S6-253753, S6-253754, S6-253791.
0.1.0
2025-10
SA6#69
Implementation of the following pCRs approved by SA6:
S6-254687, S6-254689, S6-254706, S6-254754, S6-254771, S6-254772, S6-254774, S6-254775, S6-254776, S6-254791.
0.2.0
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 1 Scope
| The present document …
The present document investigates further enhancements to the capabilities of IMS data channel (DC), specifically in the areas of ADC establishment in alerting phase, IMS capability exposure, and dynamic configuration of MDC2 endpoints.
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 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 23.228: “IP Multimedia Subsystem (IMS); Stage 2”.
[3] 3GPP TS 26.114 “Media handling and interaction”.
[4] 3GPP TS 23.203: "Policy and charging control architecture".
[5] 3GPP TS 23.503: "Policy and charging control framework for the 5G System (5GS); Stage 2".
[6] 3GPP TS 24.229: “IP multimedia call control protocol based on Session Initiation Protocol (SIP) and Session Description Protocol (SDP); Stage 3”
…
[x] <doctype> <#>[ ([up to and including]{yyyy[-mm]|V<a[.b[.c]]>}[onwards])]: "<Title>".
It is preferred that the reference to TR 21.905 be the first in the list.
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 3 Definitions of terms, symbols and abbreviations
| This clause and its three (sub) clauses are mandatory. The contents shall be shown as "void" if the TS/TR does not define any terms, symbols, or abbreviations.
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 3.1 Terms
| For the purposes of the present document, the terms given in 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 TR 21.905 [1].
Definition format (Normal)
<defined term>: <definition>.
example: text used to clarify abstract rules by applying them literally.
3.2 Symbols VOID
For the purposes of the present document, the following symbols apply:
Symbol format (EW)
<symbol> <Explanation>
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 3.2 VOID
| |
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 3.3 Abbreviations
| For the purposes of the present document, the abbreviations given in 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 TR 21.905 [1].
Abbreviation format (EW)
<ABBREVIATION> <Expansion>A2P Application to Person
ADC Application Data Channel
API Application Programming Interface
DC Data Channel
DCAR Data Channel Application Repository
DCMF Data Channel Media Function
DCMTSI Data Channel Multimedia Telephony Service for IMS
DCSF Data Channel Signalling Function
IMS IP Multimedia Subsystem
MF Media Function
MRF Media Resource Function
NEF Network Exposure Function
NF Network Function
OMA Open Mobile Alliance
P2A Person to Application
P2A2P Person to Application to Person
P2P Person to Person
|
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 4 Architectural Assumptions and Requirements
| |
de06d6f7b83561bec34bdae95ef8ccf9 | 23.700-56 | 4.1 Architectural Assumptions
| The following Architectural Assumptions are applicable to this Study:
• The IMS and IMS DC architecture defined in TS 23.228 [2] is used as the basis for any enhancements.
• An AS may be trusted or untrusted.
• The functionality of existing standardized APIs (e.g., OMA APIs) exposing MMTEL services may be used where applicable when the MMTEL service does not depend on IMS DC media.
• OMA APIs are not extended to support Data Channel and the principles outlined in TS 23.228 [2] Annex AD (IMS Subscribe/Notify framework for event monitoring).
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.