BT says no prizes yet for ‘gold standard’ RIC xApps

This week’s Open RAN Summit from Fierce Wireless focused one of its sessions entirely on the RAN Intelligent Controller (RIC), highlighting its importance in the development of open RAN services in 5G networks.

RIC is a software that leverages AI/ML to help automate the RAN and maximize network efficiency. In addition, the RIC can run applications developed by third-party specialist software providers, known as “xApps” and “rApps,” which are now available for operators to implement on their real and near-real time RICs. The question is, will RIC live up to its promise to automate and optimize the RAN, and what is the current status with xApps and rApps?

BT was one of the participants on a panel during the summit that discussed this very topic. It’s probably fair to say that the UK-based telco was, initially, not one of the most enthusiastic supporters of open RAN, certainly when compared to its peers such as Deutsche Telekom and Vodafone.

Then around 18 months ago, BT launched a trial of Nokia’s RIC in an attempt to show its commitment to open RAN, although that in turn provoked a heated debate on social media about the nature of open RAN. It also raised the question: is the RIC alone enough to constitute open RAN? Alternatively, does open RAN even need a RIC?

Richard MacKenzie, distinguished engineer in Wireless Networks at BT, conceded that there are still many grey areas about what an open RAN solution actually is. “Certainly, you can deploy open RAN without a RIC, it's just another way to form a base station. From day one in a basic deployment, if you know how to configure it, you’ve got your configuration file, there’s no reason that you have to have a RIC,” he said.

“What I suspect we will see a lot of,” he added, “are legacy base stations or base stations that are not necessarily disaggregated that might benefit from having a RIC. If you can already coordinate and optimize your open RAN deployment and you already have a RIC deployed, why not let that RIC coordinate all of the radios?”

MacKenzie also noted that he has not yet seen what he would describe as any well-established commercial RIC deployments, although he said there are “certainly lots of live trials.”

Currently, MacKenzie’s key research focus at BT labs in Adastral Park is encouraging open RAN maturity and adoption. He also engages with industry groups such as the Telecom Infra Project (TIP) and is one of the co-chairs of TIP’s RAN Intelligence & Automation (RIA) group that concentrates on the maturity and accelerated adoption of the RIC.

xApps and rApps

On the topic of the interoperability and portability of xApps and rApps between different RICs, he said this is “one of the grey areas where you could say it’s looking good already, or you could say that actually there’s a long way to go.”

For instance, xApps are not portable in the sense that they cannot be moved between different platforms without some level of integration work, he commented. The “really big challenge,” he said, is to make sure that “as we build up a portfolio of xApps and rApps, they work together.”

Within TIP, “one of the things we’ve been doing is trying to get the industry focused on a subset of use cases to make sure that we focus on the key challenges and share information, and that way as an industry we can mature as quickly as possible to commercially-ready solutions."

So far, 15 use cases have been identified. “For each use case, there may be one or a few xApps, but what we really want is a full solution that addresses all these use cases in parallel, and so we end up with a number of xApps that might even have conflicting objectives,” he said.

This in turn means that the concept of a conflict resolution xApp or rApp — that is, an xApp that effectively monitors the behavior of all the other xApps — “tends to be one of the more common solutions that we're now investigating within the industry,” he said.

MacKenzie added: “One of the reasons why we as an operator really like the RIC … is the fact that it’s not just about automation and optimization, it’s differentiation as well. So we do actually want to be in a position where we can provide a service to our customers [that is] different from our competitor operators. Part of the way we can do that is by collecting together various xApps in a different way, to combine, to produce a different service. So when we do that, we have to make sure that these xApps, which could all come from different third parties, don’t conflict with each other.”

As yet, MacKenzie said the industry is still in the “period of discovery” when it comes to conflict resolution xApps/rApps, as well as apps in general. “We’re actually putting these xApps together and we're seeing unforeseen consequences, and it’s about learning what are these consequences and … [addressing] them as well.”

TIP is also starting to give out badges, he added. “We can give out bronze badges if you’ve got a solution that satisfies the right requirements of one of our prioritized use cases. And then when you demonstrate it in a vendor-neutral lab, you can get a silver badge. And what we’re trying to give out, and we haven’t given any so far, is gold badges, which is these xApps in live deployments,” he said.