5G Network Data Analytics Functions (NWDAF)
NWDAF is a 5G SA Core network function that leverages network data analytics to generate the real-time operational intelligence driving network automation and service orchestration in the 5G.
NWDAF is responsible for processing and analyzing massive amounts of data collected from various sources within the network. According to 3GPP specifications, NWDAF is designed as a stand-alone functional entity within the 5G Core network architecture. It interacts with other network functions through standardized interfaces, such as the Nnwdaf (NWDAF Network) interface and the Nnwdaf Management (NM) interface.
The NM interface enables communication between the NWDAF and the Nnwdaf Management Function (NMF). The NMF is responsible for managing NWDAF instances, configuring NWDAF analytics policies, and monitoring NWDAF’s performance.
Network Functions supporting the data analytics architecture implement an event exposure service-based API. The NWDAF subscribes to uniquely identified events published by Network Functions to acquire the data and to analyze and expose it to subscribers.
NWDAF interacts with different entities:
- Data collection based on subscription to events provided by AMF, SMF, PCF, UDM, AF (directly or via NEF)
- Retrieval of information from data repositories (e.g. UDR via UDM for subscriber-related information)
- Retrieval of information about NFs (e.g. NRF for NF-related information)
- On demand provision of analytics to consumers
- Provision of bulked data to consumers
- Analytics and Data collection using the DCCF (Data Collection Coordination Function), including OAM data
The Nnf interface is defined for the NWDAF to request subscription to data delivery for a particular context, cancel subscription to data delivery and to request a specific report of data for a particular context.
NWDAF has three categories for analytics.
1. UE (User Equipment)
2. Service experience
3. Network Load & Performance
All three categories are further grouped per UE, groups of UEs, and slice application.
Category: Network related analytics
NWDAF provide support for two types of network performance events:
=> “num_of_ue”, measures the number of attach requests during a time window
=> “sess_succ_rate”, measures the session success rate during a time window specified in the request
NWDAF clients can subscribe to the “abnormal_behaviour” event and exception ID “unexpected_large_rate_flow” to receive periodic updates on the probability of abnormal traffic.
Category: UE Related Analytics
UE Communications: “ue_comm” refer to the number of packets and bytes exchanged in the uplink and downlink directions for each PDU session
UE Mobility: “ue_mobility” refers to the cell ID of a given UE
NWDAF deployment
Each NWDAF is identified by an analytics id and area of interest. Single NWDAF can provide multiple IDs. The Area of Interest is the geographical location that the NF belongs to.
A single instance or multiple instances of NWDAF may be deployed in a PLMN. If multiple NWDAF instances are deployed, the architecture supports deploying the NWDAF as a central NF, as a collection of distributed NFs, or as a combination of both. If multiple NWDAF instances are deployed, an NWDAF can act as an aggregate point (i.e. Aggregator NWDAF) and collect analytics information from other NWDAFs, which may have different Serving Areas, to produce the aggregated analytics (per Analytics ID), possibly with Analytics generated by itself.
NWDAF Data Analytics
3GPP has defined a group of standard functions for supporting data analytics in 5G Network Deployments:
● AnLF — Analytical Logical Function: It performs inference, derives analytics information (i.e. derives statistics and/or predictions based on Analytics Consumer request) and exposes analytics service i.e. Nnwdaf_AnalyticsSubscription or Nnwdaf_AnalyticsInfo.
● MTLF — Model Training Logical Function: It trains Machine Learning (ML) models and exposes new training services (e.g. providing trained ML model).
● DCCF — Data Collection Coordination (& Delivery) Function: It coordinates the collection and distribution of data requested by NF consumers. It prevents data sources from having to handle multiple subscriptions for the same data and send multiple notifications containing the same information due to uncoordinated requests from data consumers.
● ADRF — Analytical Data Repository Function: The ADRF offers services that enable a consumer to store and retrieve data and analytics.
● MFAF — Messaging Framework Adaptor Function: a Messaging Framework is a means by which analytics or event notifications (carrying raw data for the NWDAF to process) can be distributed around the network.
The 5G System architecture allows NWDAF and DCCF to collect data from an NWDAF with associated Nnwdaf_DataManagement services. The 5G system architecture allows MFAF to fetch data from an NWDAF with associated Nnwdaf_DataManagement service.
NWDAF — ML Model Deployment
NWDAF MTLF trains and deploys the model inference microservice. AnLF provides historical data that the Model Microservice requires for prediction. DCCF is the central point for managing all the data requests. The data transfer happen between the MFAF (Messaging Framework Adaptor Function) and ADRF (Analytical Data Repository Function). ADRF stores the historical data.
In Release 17, 3GPP has defined a new reference architecture for collecting UE data and exposing it to downstream consumers. As well as supporting generic UE data, such as service experience, the abstract building blocks are intended to be reusable in any number of different data domains.
To support the collection of data from UEs, an Application Function will receives data reports from a Data Collection Client via conventional data plane PDU Sessions and exposes events to the NWDAF using the Naf_EventExposure service-based interface defined by SA2 and specified by CT3 in TS 29.517.
Scope of the UE data collection, reporting and exposure system is further expanded in two different directions:
- Application Server (AS) instances can be an additional source of UE data. Application logs contain valuable data about UE activity that can be analyzed to extract patterns and trends.
- Third party Application Service Providers (ASPs) are valid consumers of UE events from the Data Collection AF in addition to the NWDAF. An ASP’s Event Consumer AF may wish to perform its own analysis of UE data in order to optimize its applications.
References:
TS23.288 Architecture enhancements for 5G System (5GS) to support network data analytics services
TS23.501 System architecture for the 5G System (5GS)
TS29.517 5G System; Application Function Event Exposure Service; Stage 3
TS26.531 Data Collection and Reporting; General Description and Architecture
TS26.532 Data Collection and Reporting; Protocols and Formats
