Flex-Net: A Graph Neural Network Approach to Resource Management in Flexible Duplex Networks

TL;DR

Flex-Net is a novel graph neural network designed to optimize resource management in flexible duplex networks, achieving near-optimal sum-rate performance with lower computational complexity and better scalability than traditional heuristics.

Contribution

This paper introduces Flex-Net, a GNN architecture that jointly optimizes communication direction and power in flexible duplex networks, addressing scalability and complexity issues.

Findings

Flex-Net achieves near-optimal sum-rate performance.

Flex-Net outperforms existing heuristics in scalability and complexity.

GNNs demonstrate advantages in sample complexity and generalization.

Abstract

Flexible duplex networks allow users to dynamically employ uplink and downlink channels without static time scheduling, thereby utilizing the network resources efficiently. This work investigates the sum-rate maximization of flexible duplex networks. In particular, we consider a network with pairwise-fixed communication links. Corresponding combinatorial optimization is a non-deterministic polynomial (NP)-hard without a closed-form solution. In this respect, the existing heuristics entail high computational complexity, raising a scalability issue in large networks. Motivated by the recent success of Graph Neural Networks (GNNs) in solving NP-hard wireless resource management problems, we propose a novel GNN architecture, named Flex-Net, to jointly optimize the communication direction and transmission power. The proposed GNN produces near-optimal performance meanwhile maintaining a low computational complexity compared to the most commonly used techniques. Furthermore, our numerical results shed light on the advantages of using GNNs in terms of sample complexity, scalability, and generalization capability.