Federated learning improves radio spectrum sharing with smart surfaces

Federated Multiagent Soft Actor Critic for IRS Assisted Cognitive Radio Networks under Partial Observability

Networking and Internet Architecture

Summary

Wireless communication networks need to share the available radio channels efficiently without interfering with important users. The authors developed a way for many devices to smartly adjust their signal power and the settings of special surfaces that help control signals, all while only using their own local information and without sharing sensitive data. They combined a type of artificial intelligence with federated learning, allowing devices to learn together safely and improve the network's overall performance. Their approach helps protect the main licensed users and reduces interference by coordinating devices through periodic sharing of model updates rather than raw data.

What this means in practice

  • For wireless network engineers: Improve spectrum use and interference control by coordinating device power and environment settings without sharing sensitive data.
  • For smart building automation teams: Manage multiple reflectors and transmitters to optimize wireless signals indoors while preserving privacy and minimizing interference.

Authors

Satvika Pullisani, Usama Mir, Ubaid Abbasi

Abstract

Intelligent reflecting surfaces (IRSs) can improve spectrum efficiency and interference control in cognitive radio networks (CRNs) by reconfiguring the propagation environment. This paper presents FMASAC, a federated multi-agent soft actor-critic framework for IRS-assisted CRNs under partial observability, in which distributed secondary users (SUs) jointly select their transmit power and the phase shifts of an assigned subset of IRS elements using only local observations. Primary user (PU) protection is incorporated through an interference-penalized reward, and SUs coordinate by periodically averaging model parameters at a federated server, without exchanging raw observations, channel state information, or replay data. We formulate the problem as a partially observable Markov decision process, specify the local learning and aggregation procedure, and characterize its per-round communication cost. Illustrative performance trends contrast FMASAC with centralized SAC, independent SAC, IRS-fixed multi-agent reinforcement learning, and a heuristic CRN strategy. We also outline the simulation-based evaluation and the open challenges, namely constraint handling, agent heterogeneity, and scalability, that are needed to validate the framework.