5 AI/ML management functionality and service framework

28.1053GPPArtificial Intelligence/ Machine Learning (AI/ML) managementManagement and orchestrationRelease 17TS

5.1 Functionality and service framework for ML training

An ML training Function playing the role of AI/ML training MnS producer, may consume various data for ML training purpose.

As illustrated in Figure 5.1-1 the ML training capability is provided via ML training MnS in the context of SBMA to the authorized consumer(s) by ML training MnS producer.

Figure 5.1-1: Functional overview and service framework for ML model training

The internal business logic of ML training leverages the current and historical relevant data, including those listed below to monitor the networks and/or services where relevant to the ML model, prepare the data, trigger and conduct the training:

– Performance Measurements (PM) as per 3GPP TS 28.552 [4], 3GPP TS 32.425 [5] and Key Performance Indicators (KPIs) as per 3GPP TS 28.554 [6].

– Trace/MDT/RLF/RCEF data, as per 3GPP TS 32.422 [7] and 3GPP TS 32.423 [8].

– QoE and service experience data as per 3GPP TS 28.405 [9] and 3GPP TS 28.406 [10].

– Analytics data offered by NWDAF as per 3GPP TS 23.288 [3].

– Alarm information and notifications as per 3GPP TS 28.532 [11].

– CM information and notifications.

– MDA reports from MDA MnS producers as per 3GPP TS 28.104 [2].

– Management data from non-3GPP systems.

– Other data that can be used for training.