Verizon tapped Samsung and Qualcomm for a multivendor deployment using an artificial intelligence (AI)-infused radio access network (RAN) management application running on a RAN intelligent controller (RIC), highlighting the rapidly evolving open RAN and AI ecosystems.

The new deployment uses Qualcomm’s Dragonwing RAN automation suite RIC, which supports a vendor-neutral rApps marketplace. The Qualcomm platform includes a data management layer to provide applications with insight for AI-driven RAN management.

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.

Samsung’s AI-powered Energy Saving Manager (AI-ESM) application is using that insight to help Verizon manage power consumption of its open RAN deployment. Verizon claims it has been able to produce 15% energy savings on average, “with a maximum of 35% per sector during low traffic periods in a variety of field trials.”

That power savings is significant for operators that are operating tens-of-thousands of individual 5G cell sites. An early report from MTN Consulting found carrier’s spent 5.2% of their capex in 2020 on electricity, fuel, and water.

The evidence that 5G is driving costs higher is “modest,” but the increases “serve as a good reminder that telcos will need to seek out energy efficient equipment, software, and network architectures as 5G penetration grows,” Matt Walker, chief analyst at MTN Consulting, wrote in the report.

Verizon open RAN, AI efforts keep running

Verizon has deployed more than 130,000 open RAN “capable” radios that are compatible with specifications from the O-RAN Alliance. These radios include massive multiple-input, multiple-output (MIMO) antenna technology.

The carrier last year launched an open RAN distributed antenna system (DAS) at the University of Texas in Austin and the Austin Convention Center using equipment from Samsung and CommScope running on top of its edge-focused virtualized cloud architecture. Adam Koeppe, SVP of technology planning at Verizon, at that time touted open RAN’s flexibility.

“Having the ability to pick and choose vendors based on their capabilities and knowing that they can interoperate on either side of that equation and yield high performance, high quality, low cost, whatever your goal is, for the network,” Koeppe said. “In this case, you’ve got a Samsung component and you’ve got a CommScope component in the DAS system, so that hasn’t been done before. It represents, really, the true spirit, if you will, of [open] RAN where you have two vendors in that supplier system on either ends of that.”

AI in the RAN

Analysts have also pointed to the benefits AI can bring to the deployment and management of cloud-based open RAN networks, which are more complex orchestration challenges due to the disaggregated multivendor ecosystem. The use of AI could help close performance gaps for open RAN architectures compared with legacy RAN models.

Koeppe explained that Verizon’s broader cloud-based network virtualization efforts have laid the foundation for greater AI usage.

“Where I see our evolution occurring is when you have an advanced cloud platform, as we do, you have an orchestration layer on top that we already have, and you then find ways to incorporate new AI capabilities on top of that, that's going to allow your engineers and your operators to interface differently. It's going to allow you to pull different insights out of the customer experience and help inform your optimization of the network for those experiences,” Koeppe said. “But it's all based on that foundation of having cloud-based infrastructure, deployment of cell-site software, [and an] orchestration layer running on top of that, AI becomes kind of the next stepping stone in that highly advanced network architecture that we've already deployed.”

The recently launched AI-RAN Alliance is looking to spearhead these industrywide efforts with backing from founding members including Amazon Web Services (AWS), Arm, Ericsson, Microsoft, Nokia, Samsung Electronics, SoftBank, Nvidia, DeepSig, T-Mobile US, and Northeastern University.

That group’s overriding goal is to steer the use of AI into RANs for better performance, lower operating costs, greater efficiencies, and to support new business models. This work will include using AI to improve RAN spectral efficiency, combine the two for more efficient network utilization, and deploy AI at the network edge to support new services.

The initiative recently gained leadership momentum when it announced Alex Choi as head of the group. Choi has previously served in a similar position at the O-RAN Alliance.

T-Mobile US was one of the first operators to jump on the initiative. The operator last year partnered with Nvidia, Ericsson, and Nokia to build an AI-RAN test facility in Bellevue, Washington. The carrier noted in a presentation that the goal is to integrate cloud-based RAN and AI using unified infrastructure that can scale to serve millions of mobile users at once.

“AI-RAN will enable new AI algorithms to unlock the full potential of wireless networks,” T-Mobile US’ presentation noted. “These AI algorithms would be rapidly developed with software-defined RAN, trained on AI data centers, and fine-tuned with physically accurate digital twins. This will lead to dramatic improvements in spectral and energy efficiencies.”

ABI Research’s Larbi Belkhit noted in a report that these benefits will drive future RAN innovations.

“Long term, the industry focus among both operators and vendors will continue shifting toward the AI-RAN concept, which does not rely on open interfaces for implementation and aims to address the need for better monetizing network assets at the edge,” Belkhit wrote.