NWDAF with Smart Analytics Based on Smart Data

NWDAF with Smart Analytics Based on Smart Data

Authored by Rick Fulwiler and John English (Part 2 of 2)

To reap the full benefits from automation and analytics, communications service providers (CSPs) need to operate based on a single data source. Then there’s digital transformation (DX), which also requires a single data source across the 5G radio access network (RAN), edge, and core. The problem is, data at CSPs is nearly always siloed. And without a single data source, different functional network teams operate on different truths, so to speak. To arrive at a single source of truth, however, CSPs need to find a single source of data that is vendor-, technology-, and cloud-agnostic.

One such option is NETSCOUT Smart Data. The foundation of NETSCOUT’s service intelligence is smart data, which is derived from direct access to the wired data source—namely, the IP packets that traverse the 5G stand-alone (SA) service-based architecture. NETSCOUT Automated Analytics takes the raw power of NETSCOUT Smart Data to the next level by recognizing business impact across any CSP network. By injecting specific domain knowledge into highly tuned artificial intelligence (AI) and machine learning (ML) algorithms, Automated Analytics drives a powerful workflow automation engine that can quickly isolate business impact across multiple domains, from access to core, automatically providing actionable business intelligence that otherwise might have taken days to if not weeks to isolate—and in a visually intuitive display.

The combined capabilities of NETSCOUT Smart Data and Automated Analytics make up the cornerstone of NETSCOUT's ability to provide advanced 5G analytics to drive closed-loop automation use cases as defined within the 3GPP network data analytics function (NWDAF) specification. Automated Analytics goes beyond providing a list of fixed use cases: It provides an environment that is pliable to how the carrier works and what drives the carrier’s business—which is where the line of business in the CSP organization needs to help define use cases that impact business.

Our Automated Analytics applications take advantage of NETSCOUT’s powerful Smart Data framework, which is a multi-technology, multidimensional, subscriber-based data set that contains rich application and network metrics derived directly from packets: the ultimate source of truth.

Automated Analytics Supports 3GPP NWDAF Use Cases

NETSCOUT’s Automated Analytics product suite is designed to help resolve the increasing network complexity added by 5G. Automated Analytics uses Smart Data to simplify the underlying changes in the 5G network, including virtualization and cloudification of the core, edge computing, network slicing, and the RAN. It executes millions of investigative iterations regardless of network deployment to find the issues with the highest impact and their underlying causality.

Automated Analytics can address the use cases outlined in 3GPP TR23.791 for formula-based AI/ML analytics; with this solution, we expect impacted user experience detection/prediction cases will be important. There is great value in any measurement that can assist in closed-loop automation use cases. To be successful, each of these NWDAF use cases requires the highest level of data resolution, which NETSCOUT can provide. Our virtual data acquisition engines can take in RAN metrics and core-control and user-plane packet-level data without burdening the carrier, because user-plane traffic scales in response to traffic increases. From there, we apply our domain knowledge, which is built into our AI algorithms, to drive specific use cases.

Automated Analytics increases network and operational engineering efficiency by removing the data overload engineers face and providing focus on issues with the highest impact to subscribers and services. It automates the discovery of service-impacting events by removing many of the time-consuming, manual steps currently employed


NETSCOUT believes that carriers need a platform that is pliable to how they work and what drives their business. Once line-of-business executives in a carrier’s organization help to define business use cases, NETSCOUT overlays more than 30 years of telecom experience into business analytics checklists that sift through terabytes of data for unique outliers to determine network function loading, device mobility, subscriber user experience, and network-related Information such as quality of service (QoS), sustainability, and RAN congestion, just to name a few.

Automated Analytics improves customer experience by decreasing trouble-ticket resolution times; it also helps network operators provide increased service quality. Chasing individual trouble tickets is tedious, time-consuming, and requires customer involvement in the quality process, whereas automated workflows will lower the need for customer interaction when troubleshooting service issues.

Which NWDAF Cases?

5G slicing can benefit from NWDAF. Despite 3GPP specifications focusing on the , carriers likely begin with the RAN domain, where NWDAF data can drive dynamic radio topology, power control settings on mm-Wave, employ interference mitigation techniques, add energy efficiency improvements, identify RAN SLA slice violations, and so on. From there, the mobile operator can introduce additional uses cases, bringing in dynamic resource allocation for areas such as multi-edge computing (MEC), core, and QoS optimization. It is still early going with network slicing implementation, but low-latency network slices will need the most immediate real-time adjustments with live network operation.

Although we expect it to be a very important use case for NWDAF, carriers will likely first implement static network slicing rather than the nirvana of dynamic or elastic slicing on demand. The RAN is the first point in the mobile network to address slicing issues. By identifying cell congestion and notifying the Policy Control Function (PCF) of that condition, it is able it to throttle traffic to mitigate the congestion.

From a service assurance perspective, NETSCOUT Smart Data sees the slice identifiers—slice/service type (SST), enhanced mobile broadband (eMMB), ultra-reliable low-latency communications (URLLC), mobile Internet of Things (mIoT), and video transmitter (Vtx), with the optional slice differentiator ()—that differentiate between similar slices. This resolution allows Smart Data to report on the health and well-being of each slice as identified by these markers. Being able to create events based upon any service-level violation that might occur related to each slice essentially means NETSCOUT provides tracking slices as a service.

An area in which the 3GPP specifications are silent is the role NWDAF could play in helping to detect anomalies that might be related to security threats. For example, when many IoT devices become infected with malware—and given that a key tenant of 5G is its high data rates—IoT clusters could create a massive distributed denial-of-service (DDoS) attack or even signaling storms through the network. This type of bandwidth consumption anomaly is something NETSCOUT’s visibility capabilities can detect. When suddenly an IoT device that was producing a few megabytes starts to produce gigabytes per hour, that is something that needs attention. Another example would be a user equipment (UE) device sending traffic that does not use its allocated IP address. This is a key in the battle against the DDoS attack vector leveraging reflection amplification.


The NWDAF needs a solution that can deliver the required visibility and actionable business intelligence at the scale, complexity, and pace that the modern IT networks are operating. In a multivendor environment, carriers will need a vendor-agnostic solution such as that provided by NETSCOUT. Our comprehensive end-through-end view with Adaptive Service Intelligence (ASI) Smart Data allows us to address the 3GPP NWDAF 5G use cases and much more at each functional area of the network (RAN, core, edge, and network slice), as well as determine network, service, and device-behavioral anomalies that impact user experience. Our customer-first approach propels our proactive stance to help our customers solve their most pressing business use cases, enabling them to ensure real ROI and productivity gains and move toward predictive and prescriptive solutions with automation.

Learn more about NETSCOUT's 5G solutions.