FedCritic: Serverless Federated Critic Learning-based Resource Allocation for Multi-Cell OFDMA in 6G

2026-05-20

AI Summary

This work addresses the challenge of distributed downlink resource allocation in ultra-dense 6G networks, where multi-cell OFDMA systems suffer from strong interference coupling and long-term QoS constraints. To tackle this, the authors propose FedCritic, a serverless federated multi-agent actor-critic framework. FedCritic introduces a decentralized federated critic mechanism that leverages a lightweight gossip protocol over an interference graph to synchronize critic parameters, eliminating the reliance on centralized critics and joint trajectory aggregation inherent in conventional CTDE approaches. Additionally, it incorporates virtual queue deficit weights to handle long-term QoS requirements. Experimental results demonstrate that under full frequency reuse (reuse factor of 1), FedCritic significantly improves average SINR, cell-edge rates, network throughput, and fairness, while achieving more stable training and lower communication overhead.

Abstract

In sixth-generation (6G) ultra-dense networks, aggressive frequency reuse amplifies inter-cell interference (ICI), making multi-cell orthogonal frequency-division multiple access (OFDMA) scheduling and power control strongly coupled across neighboring cells. We study distributed downlink resource management -- joint subcarrier scheduling and power allocation -- under interference coupling and long-term per-user quality-of-service (QoS) minimum-rate constraints. By using virtual-queue deficit weights to enforce long-term QoS, we develop FedCritic, a serverless federated multi-agent actor-critic framework with decentralized execution. Unlike centralized training with decentralized execution (CTDE) approaches that require centralized critic learning and joint trajectory aggregation, FedCritic federates the critic through lightweight gossip-based parameter averaging over the interference graph, enabling stable value estimation without a central coordinator while keeping policies local. Simulations in an interference-rich reuse-1 setting show that FedCritic improves mean signal-to-interference-plus-noise ratio (SINR) and cell-edge rate, increases network-wide average sum-rate and fairness relative to non-coordinated and CTDE baselines, and achieves more stable training with lower coordination overhead.

Problem

• inter-cell interference
• OFDMA
• resource allocation
• QoS constraints
• 6G ultra-dense networks

Innovation

• federated critic learning
• serverless multi-agent RL
• interference coordination
• OFDMA resource allocation
• gossip-based averaging