QoS-aware State-Augmented Learnable Algorithm for Wireless Coexistence Parameter Management

TL;DR

This paper presents QaSAL-CPM, a novel reinforcement learning framework that dynamically manages coexistence parameters in wireless networks, ensuring QoS and robustness in heterogeneous 5G NR-U and Wi-Fi environments.

Contribution

It introduces a state-augmented constrained reinforcement learning algorithm for real-time wireless coexistence management, embedding dual variables into the agent's observations.

Findings

QaSAL-CPM achieves reliable QoS compliance.

It improves policy robustness across various densities.

The framework is scalable and adaptive.

Abstract

Efficient and fair coexistence in unlicensed spectrum is essential to support heterogeneous networks such as 5G NR-U and Wi-Fi, which often contend for shared wireless resources. We introduce a general framework for wireless Coexistence Parameter Management (CPM) based on state-augmented constrained reinforcement learning. We propose a novel algorithm, QaSAL-CPM, which incorporates state-augmentation by embedding the dual variables in the constrained optimization formulation directly into the agent's observation space. This method enables the agent to respond to constraint violations in real time while continuing to optimize a primary performance objective. Through extensive simulations of 5G NR-U and Wi-Fi coexistence scenarios, we show that QaSAL-CPM achieves reliable QoS compliance and improved policy robustness across various transmitter densities compared to previous approaches. The proposed framework offers a scalable and adaptive solution for real-time coexistence optimization in next-generation wireless networks.