3GPP Slices Away at Machine Learning

Network slicing -- or the ability to run different virtual networks for specific users on the same physical network -- is one of the most promising features of 5G, and development of it is becoming closer to reality with help from the standards world.

The 3GPP is working on a network data analytics function (NWDAF) that uses machine learning to enable 5G operators to monitor the status of a network slice or third-party application performance in the 5G core. Sprint Senior Technology Strategist Serge Manning discussed the early-stage work that's underway at the Zero Touch & Carrier Automation Congress in Madrid this week.

The analytics platform would make use of any data in the network core to do things like monitor the status of the load of a network slice, understand behavior of mobile devices and monitor application performance, Manning said, according to Light Reading's Iain Morris.

This is important because network slicing alone isn't necessarily enough to differentiate service if they are static. It's the ability to have highly automated network slices that can be changed on the fly, even created on their own by third-party applications, that makes it exciting. Analysys Mason Analyst Caroline Chappell warned, "If operators don't automate, they will be providing capacity-based slices that are relatively large and static and undifferentiated and certainly not on a per-customer basis."

Read Morris's full report, including the debate over whether standards bodies are moving fast enough to realize operators' 5G ambitions, over on Light Reading: 3GPP Preps Machine Learning in 5G Core.

— Sarah Thomas, Contributing Editor, Telco Transformation