The Agentic RAN Management Service is designed for human-in-the-loop adoption on the path to AI-native 6G networks
During the recent Mobile World Congress, Qualcomm showcased AI-powered solutions for radio access network (RAN) management that are delivering real-world results today while setting the stage for increasing levels of closed-loop autonomous operations as 6G takes shape. With its Agentic RAN Management Service, Qualcomm has packaged a set of agents capable of delivering end-to-end RAN operations that can materially impact operators’ operational expenses.
Qualcomm Vice President of Engineering Dan Weil explained that traditional approaches to RAN automation involve having engineers operate and manage algorithms that ingest network telemetry data that can be used to inform an optimization or remediation action. “What we’re trying to do today is driving a significant change,” Weil said. “The key aspect here is to build an end-to-end learning system that is able to operate completely autonomously.”
He described the agentic workflow as including agents that monitor the network to identify problems while simultaneously monitoring user experience from a device standpoint. As problems are identified and classified, they’re communicated to another agentic layer that creates a chain of events resulting in an end-to-end closed-loop autonomous cycle. Additionally, an overseer agent that is able to reason, plan, execute and supervise the activities of other agents.
“As this is an evolution, we have built an entry point for human-in-the-loop in different cycles of the change,” Weil said. “This is very important to get the safety and the gradual growth.” He broke down the value proposition as twofold: “One is driving customer experience” by combining network and user experience data. Second, “We’re targeting opex here up to [a] 40% reduction. Our vision, when the network reaches this autonomous level…we’ll be able to operate the network [with] a 40% lower opex cost to the carrier.”
While the current industry focus is on expanding the reach and performance of 5G, including the ongoing transition to a 5G Standalone architecture and adoption of 5G-Advanced features, standards bodies are already working on the technical specifications of future 6G networks. The vision of AI-native 6G considers AI deeply embedded at every layer of the network stack. But, as Weil pointed out, AI-enabled automation tooling like what Qualcomm demoed at MWC is capable of driving meaningful change today.
“A lot of these concepts are now being discussed in the context of 6G,” he said. “However, we’re building those capabilities into the 5G network. What we’re building today…is targeted to operate through 5G, LTE and will be 6G-ready.”
Beyond RAN Automation, the Agentic RAN Management Service complements Qualcomm’s additional RAN AI tools introduced at MWC for use with their Open RAN RU and DU Platforms, that do things like improve uplink throughput and performance in dynamic radio conditions, downlink beamforming channel predictions and massive MIMO radio factory calibration. And, again, these are clear and present capabilities applicable now on deployed platforms, not only lab demos that could lead to future products.
Discussing the larger context surrounding Qualcomm’s AI-enabled RAN solutions, EVP and GM of Technology Planning, Edge Solutions and Data Center Durga Malladi wrote: “As networks grow more complex and more software driven, the way we operate them must fundamentally change. 6G will demand networks that are autonomous by design, able to support AI-native services at scale with high reliability and efficiency. The path there does not start in the future. It starts by deploying AI where it can already make a difference and evolving operations step-by-step toward autonomy.”
Malladi said Qualcomm’s production-ready RAN AI solutions are delivering “near-term impact, not just long-term roadmaps. This reflects a broader philosophy we have held across wireless generations: real progress comes from translating research into deployable systems, and from building platforms that let the ecosystem move forward together.”
