Tech-invite  3GPPspace  IETFspace
21  22  23  24  25  26  27  28  29  31  32  33  34  35  36  37  38  4‑5x

Content for  TS 23.288  Word version:  19.0.0

Top   Top   Up   Prev   Next
1…   4…   5…   5A…   6…   6.1.3   6.1.4…   6.1.4.4…   6.1.5…   6.1A…   6.1B…   6.1B.2.3…   6.1C   6.2…   6.2.3…   6.2.6…   6.2.6.2   6.2.6.3…   6.2.6.3.3   6.2.6.3.4   6.2.6.3.5   6.2.6.3.6…   6.2.7…   6.2.8…   6.2.9…   6.2.13…   6.2A…   6.2B…   6.2B.3   6.2B.4…   6.2C…   6.2D…   6.2E…   6.2F…   6.3…   6.4…   6.5…   6.6…   6.7…   6.7.3…   6.7.4…   6.7.5…   6.8…   6.9…   6.10…   6.11…   6.12…   6.13…   6.14…   6.16…   6.17…   6.18…   6.19…   6.20…   6.21…   7…   7.4…   7.7…   7.9…   8…   9…   10… 
7.9 Nnwdaf_MLModelMonitor Service  7.9.1 General  7.9.2 Nnwdaf_MLModelMonitor_Subscribe service operation  7.9.3 Nnwdaf_MLModelMonitor_Unsubscribe service operation  7.9.4 Nnwdaf_MLModelMonitor_Notify service operation  7.9.5 Nnwdaf_MLModelMonitor_Register  7.9.6 Nnwdaf_MLModelMonitor_Deregister  7.10 Nnwdaf_MLModelTraining Service  7.10.1 General  7.10.2 Nnwdaf_MLModelTraining_Subscribe service operation  7.10.3 Nnwdaf_MLModelTraining_Unsubscribe service operation  7.10.4 Nnwdaf_MLModelTraining_Notify service operation  7.11 Nnwdaf_MLModelTrainingInfo Service  7.11.1 General  7.11.2 Nnwdaf_MLModelTrainingInfo_Request service operation 
...

 

7.9  Nnwdaf_MLModelMonitor Service |R18|  p. 302

7.9.1  General  p. 302

Service Description:
This service enables the consumer to subscribe/unsubscribe for ML Model accuracy (i.e. Analytics accuracy for an ML Model as described in clause 6.2E.3.3) information monitored. The service can additionally provide Analytics Feedback Information for the analytics generated by an NWDAF (i.e. NWDAF containing AnLF). The service also enables the NWDAF containing AnLF registers the use and monitoring capability for an ML Model into the model provider NWDAF, i.e. NWDAF containing MTLF.

7.9.2  Nnwdaf_MLModelMonitor_Subscribe service operation  p. 302

Service operation name:
Nnwdaf_MLModelMonitor_Subscribe
Description:
Subscribes to NWDAF for the monitored ML Model accuracy (i.e. Analytics accuracy for an ML Model as described in clause 6.2E.3.3) information and Analytics Feedback Information for the analytics generated by the NWDAF with specific parameters.
Inputs, Required:
(set of) Unique ML Model identifier(s), Notification Target Address (+ Notification Correlation ID).
Inputs, Optional:
Subscription Correlation ID (in the case of modification of the ML Model monitor subscription), desired Accuracy metrics to indicate the metrics to calculate the accuracy information, reporting period to indicate the reporting periodicity in which the monitored ML Model Accuracy Information shall be reported, Accuracy reporting threshold to indicate the reporting condition across which the accuracy information shall be reported, Analytics ID, Target of Analytics Reporting and Analytics filter.
Analytics ID, Target of Analytics Reporting and Analytics filter for each ML Model identifier should reflect the corresponding information received in the Nnwdaf_MLModelMonitor_Register request from the NWDAF containing AnLF to which the current NWDAF containing MTLF subscribes.
Outputs Required:
When the subscription is accepted: Subscription Correlation ID (required for management of this subscription), Expiry time (required if the subscription can be expired based on the operator's policy).
Outputs, Optional:
None.
Up

7.9.3  Nnwdaf_MLModelMonitor_Unsubscribe service operation  p. 302

Service operation name:
Nnwdaf_MLModelMonitor_Unsubscribe
Description:
The NF consumer unsubscribes to the NWDAF for the monitored ML Model accuracy (i.e. Analytics accuracy for an ML Model as described in clause 6.2E.3.3) information and Analytics Feedback Information for the analytics generated by the NWDAF.
Inputs, Required:
Subscription Correlation ID.
Outputs, Required:
Operation execution result indication.
Outputs, Optional:
None.

7.9.4  Nnwdaf_MLModelMonitor_Notify service operation  p. 302

Service operation name:
Nnwdaf_MLModelMonitor_Notify
Description:
NWDAF notifies the monitored ML Model accuracy (i.e. Analytics accuracy for an ML Model as described in clause 6.2E.3.3) information and Analytics Feedback Information for the analytics generated by the NWDAF to the consumer instance which has subscribed to the specific NWDAF service.
Inputs, Required:
Notification Correlation Information, at least one of the following:
  • the monitored ML Model Accuracy Information which includes:
  • Unique ML Model identifier;
  • Monitoring interval: time interval during which the ML Model Accuracy Monitoring was conducted;
  • Monitored Analytics metrics value of the ML Model and a deviation value which indicates the deviation of the predictions generated using the ML Model(s) from the ground truth data;
  • Number of inferences that were performed during the monitoring interval;
  • Used Accuracy metrics (as requested in Subscribe service operation).
  • Analytics Feedback Information: indicates that the consumer NF of the analytics generated by the provisioned ML Model has taken an action(s) influenced by the analytics and includes the following parameter(s):
    • Corresponding Analytics ID(s) which has been used for taking an action(s);
    • Corresponding ML Model identifier(s) which has been used for generating Analytics;
    • Indication whether the action will affect on ground truth data (if available);
    • Time stamp(s) when the action(s) are taken.
Inputs, Optional:
  • Input data used for inferencing indicated by DataSetTag with ADRF ID when the prediction generated from the ML Model is not correct (which can be used by the NWDAF containing MTLF for possible ML Model retraining);
  • An indication that the analytics metrics value of the ML Model does not meet the requirement of accuracy for the ML Model.
  • Analytics ID;
  • Target of Analytics Reporting;
  • Analytics filter for which the model is used for.
Outputs, Required:
Operation execution result indication.
Outputs, Optional:
None.
Up

7.9.5  Nnwdaf_MLModelMonitor_Register  p. 303

Service operation name:
Nnwdaf_MLModelMonitor_Register
Description:
The consumer registers the use and monitoring capability for an ML Model at an NWDAF containing MTLF.
Inputs, Required:
Consumer NF ID, Unique ML Model identifier.
Inputs, Optional:
Endpoint address of the Nnwdaf_MLModelMonitor_Subscribe service operation, ML Model accuracy transfer indication as defined in clause 6.2E.3.2, Analytics ID, Target of Analytics Reporting and the Analytics filter for which the model is used for.
Outputs, Required:
ML Model monitoring registration ID.
Outputs, Optional:
None.

7.9.6  Nnwdaf_MLModelMonitor_Deregister  p. 303

Service operation name:
Nnwdaf_MLModelMonitor_Deregister
Description:
The consumer deregisters, from an NWDAF containing MTLF, a previous MLModelMonitor registration, e.g. when the consumer is no longer using or monitoring the accuracy of the analytics generated using the ML Model.
Inputs, Required:
ML Model monitoring registration ID.
Inputs, Optional:
A termination indication, a termination cause, the NWDAF containing AnLF NF ID of the target NWDAF (in the case that the termination cause is due to analytics transfer).
Outputs, Required:
None.
Outputs, Optional:
None.

7.10  Nnwdaf_MLModelTraining Service |R18|  p. 304

7.10.1  General  p. 304

Service Description:
This service enables the consumer to subscribe/unsubscribe/notify/modify for ML Model training.
When used for Federated Learning, this service enables FL server NWDAF to enable Federated Learning while providing global ML Model information to FL Client NWDAF and getting local ML Model information and status report of FL training as defined in clause 6.2C.2.3 from the FL Client NWDAF.
This service may also be used by the consumer (i.e. FL Server NWDAF) to check if the service provider (i.e. FL Client NWDAF) can meet the ML Model training requirement as described in clause 6.2F.1.
This service may also be used by the consumer (i.e. FL Server NWDAF) to request the service provider (i.e. FL Client NWDAF) to calculate and provide Model Accuracy of the global ML Model as described in clause 6.2F.1.
Up

7.10.2  Nnwdaf_MLModelTraining_Subscribe service operation  p. 304

Service operation name:
Nnwdaf_MLModelTraining_Subscribe
Description:
Subscribes to NWDAF ML Model training with specific parameters.
Inputs, Required:
  • Analytics ID as defined in Table 7.1-2;
  • ML Model Interoperability information;
  • Notification Target Address (+ Notification Correlation ID).
Inputs, Optional:
  • ML Model identifier: identifies the provided ML Model.
  • ML Model Information (as defined in clause 6.2A.2);
  • ML Model file;
  • Subscription Correlation ID (in the case of modification of the ML Model Training subscription);
  • ML Training Information, i.e. data availability requirement, time availability requirement.
  • ML Preparation Flag;
  • ML Model Accuracy Check Flag;
  • ML Correlation ID;
  • Training Filter Information;
  • Target of Training Reporting;
  • Training Reporting Information as defined in clause 6.2F.2;
  • Use case context;
  • Iteration round ID;
  • Expiry time.
  • Indication of skipping the current FL round.
Outputs Required:
When the request is accepted: Subscription Correlation ID (required for management of this subscription). When the request is not accepted, an error response with cause code (e.g. NWDAF does not meet the ML training requirements, ML training is not complete, NWDAF overload, not available for the FL process anymore, etc.).
Outputs, Optional:
ML Correlation ID (e.g. confirm of the subscription for this FL process).
Up

7.10.3  Nnwdaf_MLModelTraining_Unsubscribe service operation  p. 305

Service operation name:
Nnwdaf_MLModelTraining_Unsubscribe
Description:
Terminate NWDAF ML Model training.
Inputs, Required:
Subscription Correlation ID.
Inputs, Optional:
None.
Outputs, Required:
Operation execution result indication.
Outputs, Optional:
Cause code (e.g. FL Client NWDAF is unselected by the FL Server NWDAF for the FL process, or the FL process is suspended or finished, etc.). Final aggregated ML Model information (if FL has finished) or updated aggregated ML Model information (if FL is suspended).

7.10.4  Nnwdaf_MLModelTraining_Notify service operation  p. 305

Service operation name:
Nnwdaf_MLModelTraining_Notify
Description:
NWDAF notifies the consumer instance of the trained ML Model that has subscribed to the specific NWDAF service. The NWDAF can also use this service to indicate to consumer it will terminate the ML Model training.
Inputs, Required:
  • Notification Correlation Information: this parameter indicates the Notification Correlation ID that has been assigned by the consumer during ML Model training.
Inputs, Optional:
  • Set of the tuple (Analytics ID, ML Model Information as defined in clause 6.2F.2;
  • ML Correlation ID, when for Federated Learning;
  • Corresponding Use case context;
  • Termination Request: this parameter indicates that NWDAF requests to terminate the ML Model training, i.e. NWDAF will not provide further notifications related to this request, with cause code (e.g. NWDAF overload, not available for the FL process anymore, etc.);
  • ML Model identifier: this parameter identifies the provisioned ML Model;
  • Global ML Model Accuracy information: The model accuracy of the global ML Model, which is calculate by the FL Client NWDAF using the local training data as the testing dataset;
  • Status report of FL training: local ML Model accuracy information and Training Input Data Information (e.g. areas covered by the data set, sampling ratio, maximum/minimum of value of each dimension, etc.), which are generated by the FL Client NWDAF during FL procedure;
  • Delay Event Notification: as defined in clause 6.2F.2;
  • Iteration round ID.
Outputs, Required:
Operation execution result indication.
Outputs, Optional:
None.
Up

7.11  Nnwdaf_MLModelTrainingInfo Service |R18|  p. 306

7.11.1  General  p. 306

Service Description:
This service enables the consumer to request for the information about ML Model training based on the ML Model file or ML Model information as described in clause 6.2F.2 provided by the consumer.
When used for Federated Learning, this service enables FL server NWDAF to enable Federated Learning while providing global ML Model information to FL Client NWDAF and getting local ML Model information from the FL Client NWDAF.
Up

7.11.2  Nnwdaf_MLModelTrainingInfo_Request service operation  p. 306

Service operation name:
Nnwdaf_MLModelTrainingInfo_Request
Description:
Request information about NWDAF ML Model training with specific parameters.
Inputs, Required:
  • Analytics ID as defined in Table 7.1-2.
  • ML Model Interoperability information.
Inputs, Optional:
  • ML Model identifier: identifies the provided ML Model.
  • ML Model Information (as defined in clause 6.2A.2).
  • ML Model file.
  • ML Training Information (i.e. data availability requirement, time availability requirement).
  • Training Reporting Information as defined in clause 6.2F.2.
  • ML Preparation Flag.
  • ML Model Accuracy Check Flag.
  • ML Correlation ID.
  • Termination Request, when terminating the Federated Learning identified by the ML Correlation ID and optionally indicating the reason, e.g. FL Client NWDAF is unselected by the FL Server NWDAF for the FL process, or the FL process is suspended, etc.
  • Training Filter Information.
  • Target of Training Reporting.
  • Use case context.
  • Indication of skipping the current FL round.
Outputs Required:
When the request is accepted: Operation execution result indication. When the request is not accepted, an error response with cause code (e.g. NWDAF does not meet the ML training requirements, ML training is not complete, NWDAF overload, not available for the FL process anymore, etc.).
Outputs, Optional:
  • ML Model identifier.
  • Set of the tuple (Analytics ID, ML Model Information (as defined in clause 6.2A.2)).
  • ML Correlation ID, when for Federated Learning.
  • Corresponding Use case context.
  • Global ML Model Accuracy information: The model accuracy of the global ML Model, which is calculate by the FL Client NWDAF using the local training data as the testing dataset.
  • Status report of FL training: local ML Model Accuracy Information and Training Input Data Information (e.g. areas covered by the data set, sampling ratio, maximum/minimum of value of each dimension of data, etc.), which are generated by the FL Client NWDAF during FL procedure.
  • Delay Event Notification as defined in clause 6.2F.2.
Up

Up   Top   ToC