AT&T and Nokia completed a trial of the vendor’s near real-time radio access network (RAN) intelligent controller (RIC) running xApps using the rapidly evolving E2 interface.

The trial used Nokia’s RIC platform running on its AirScale base station on AT&T’s network. The near real-time xApps used E2SM (service model) policy services for dynamic RAN optimization.

This E2 work basically taps into the near real-time xApps running in a RIC to monitor and optimize an operator’s RAN deployment – typically either a virtualized RAN (vRAN) or open RAN – and across different spectrum bands. This in turn allows the operator to support more stringent service-level agreements (SLAs) and private network deployments that can generate more revenues.

Nokia touted use cases like advanced traffic steering xApps to better utilize an operator’s network capacity and anemology detection xApps that use machine learning (ML) to detect and classify RAN irregularities.

“The results of the trial demonstrated that E2 interface allows RAN policies to be updated several orders of magnitude faster than through legacy OAM interfaces, and thus unlocking new methods for RAN optimization,” Robert Soni, VP for RAN technology at AT&T, noted in a statement.

The RIC operates in the middle of an open RAN deployment as a kind of linkage to the wider network. It allows operators to deploy xApps and rApps that then allow operators to design and control RAN functions, providing administrative RAN sovereignty over functions that are typically implemented as proprietary features on base stations.

These “apps” are microservice-based applications operating in near-real time (xApps) and non-real time (rApps) that provide an operator with more control over their open RAN environments.

RIC work on the E2 interface has been increasing in recent months.

Yousef Khalidi, corporate VP of Azure for Operators at Microsoft, explained in a blog post late last year that the hyperscaler is working to extend the E2 interface so that it can dynamically extract detailed internal data and real-time telemetric data from live RAN software to provide more detailed RAN control for operators. The Microsoft RAN analytics and control framework would allow third-party RAN vendors to run specific applications that can extract this data for the operator without interfering with the normal RAN operation.

“With this technology, together with detailed platform telemetry, operators can achieve better network monitoring and performance optimization for their 5G networks, and enable new [artificial intelligence], analytics, and automation capabilities that were not possible before,” Khalidi wrote.

Microsoft has been working with Capgemini and Intel to develop a prototype of the system using Capgemini’s 5G RAN and Intel’s FlexRAN reference software.

That work has found the ability to produce better power modulation for a RAN deployment as the framework can better react to network load requirements. It can also better scrape data from the network to determine interference or performance issues for more precise remediation.

Brandon Larson, SVP and GM of Mavenir’s Multimedia Business Unit, recently told SDxCentral in an interview that the vendor is pushing deeper into the E2 interface that coincides with the xApps layer. This is the layer that interfaces with network data and KPIs to support specific use cases.

“We probably wrote about 70% of the spec that’s out there in the O-RAN standards,” Larson added.

The latest work between AT&T and Nokia builds on a partnership that was initially announced in early 2019. That push was targeted at developing an O-RAN Alliance compliant software platform for the RIC to support a set of functions and interfaces that allow for easier integration through policy-driven closed loop automation and more flexible deployments and programmability within the RAN.

The pair the following year conducted a limited live trial of their jointly developed RIC on AT&T’s commercial 5G millimeter-wave (mmWave) network in New York City. The apps were designed to improve spectrum efficiency and provide location- and use case-specific customizations.