DRL-Based AoI Minimization for RSMA in Finite-Blocklength MU-MISO

2026-10-08

AI Summary

This study addresses the prohibitive computational complexity of jointly optimizing Rate-Splitting Multiple Access (RSMA) parameters in finite blocklength multi-user networks by proposing a deep reinforcement learning-based framework, termed RSMA-RL. Targeting MU-MISO broadcast channels, the proposed method integrates RSMA with finite blocklength coding theory and employs an Actor-Critic algorithm to reformulate the high-dimensional non-convex optimization as a sequential decision-making process, dynamically optimizing resource allocation to minimize Age of Information (AoI). Experimental results demonstrate that RSMA-RL significantly reduces AoI under low signal-to-noise ratio (SNR) and short blocklength conditions while achieving performance comparable to baseline methods at high SNR. Furthermore, its single forward-pass inference incurs minimal computational overhead, effectively enhancing information freshness guarantees for short-packet transmissions.

Abstract

This letter investigates age-of-information (AoI) minimization in multi-user wireless networks operating in the finite-blocklength (FBL) regime, which is critical for low-latency transmission of short state-update packets. While rate-splitting multiple access (RSMA) provides a powerful and flexible framework for interference management in multi-user FBL systems, the joint optimization of its parameters, such as precoding vectors, power allocation, and rate-splitting ratios, to guarantee information freshness results in analytically intractable complexity. To address this challenge, we propose an actor--critic deep reinforcement learning (DRL) framework to learn dynamic resource-allocation policies in multi-user multiple-input single-output (MU-MISO) broadcast channels. Simulation results show that the proposed RSMA-RL framework achieves consistently lower AoI than the state-of-the-art benchmarks, with substantial gains observed at low signal-to-noise ratio (SNR) and short blocklengths, while matching benchmark performance at high SNR with significantly lower online complexity via a single neural-network forward pass at execution.

Problem

• Age-of-Information
• Finite-Blocklength
• Rate-Splitting Multiple Access
• MU-MISO

Innovation

• Age of Information (AoI)
• Rate-Splitting Multiple Access (RSMA)
• Finite-Blocklength (FBL)
• Deep Reinforcement Learning (DRL)
• MU-MISO