AI agents control and workload share compute in modern radio networks
Control-Compute Governance in Agentic AI-RAN: AI Agents as Both Controllers and Workloads
Networking and Internet Architecture
Summary
Modern radio networks use smart AI helpers based on large language models to manage communications. These helpers share computing power with the core radio tasks, but this can cause delays because they compete for resources. The authors propose a set of rules called a control-compute contract that helps balance this competition. Their tests show the contract can reduce slowdowns while keeping faster control decisions.
What this means in practice
- •For network operators: Manage AI agents and radio processing on shared compute to maintain communication speed and control responsiveness.
- •For infrastructure providers: Design and deploy coordinated compute resource allocation protocols that ensure AI-driven control does not slow down physical radio workloads.
Authors
Le Xia, Rose Qingyang Hu, Yuan Zhou, Haijian Sun
Abstract
Recent advances in artificial intelligence-radio access network (AI-RAN) are placing large language model (LLM)-driven agents within the open RAN (O-RAN) control hierarchy. A promising deployment for this agentic AI-RAN co-locates LLM inference with physical-layer (PHY) communication processing over the same accelerated compute pool for infrastructure sharing, data locality, and low control latency. However, this co-location induces bidirectional control-compute coupling, as the agent competes with PHY processing for compute, while its reasoned gNB control actions may alter future PHY workload and hence the compute available to its next inference. To this end, this article proposes the control-compute contract, a coordination protocol linking AI-RAN workload governance with O-RAN control execution. Specifically, we first expose direct compute contention on an over-the-air (OTA) O-RAN testbed, where continuous LLM inference increases the PHY decoding time approximately eightfold. We then formulate the contract as three protocol rules with operator-requirement protection, and map it onto O-RAN and AI-RAN functions as a five-stage workflow. Afterward, an OTA-calibrated case study of simulated uplink gNB control indicates that the contract keeps the proportion of slots violating the PHY time budget below $0.9\%$, against $20.7\%$ under ungoverned co-location, while achieving faster gNB control than a fixed compute reservation benchmark. Finally, we identify several open issues and outlooks for its practical deployment.