Unfolding WMMSE using Graph Neural Networks for Efficient Power Allocation

TL;DR

This paper introduces UWMMSE, a neural network architecture based on unfolding the WMMSE algorithm and utilizing GNNs for efficient, adaptable power allocation in wireless networks, demonstrating strong performance and generalizability.

Contribution

The paper proposes a novel hybrid neural network architecture that combines algorithm unfolding with GNNs for power allocation, enhancing adaptability and performance.

Findings

UWMMSE outperforms classical methods in simulations.

The architecture is permutation equivariant, enabling generalization.

Robust to hyper-parameter variations and scalable to different network sizes.

Abstract

We study the problem of optimal power allocation in a single-hop ad hoc wireless network. In solving this problem, we depart from classical purely model-based approaches and propose a hybrid method that retains key modeling elements in conjunction with data-driven components. More precisely, we put forth a neural network architecture inspired by the algorithmic unfolding of the iterative weighted minimum mean squared error (WMMSE) method, that we denote by unfolded WMMSE (UWMMSE). The learnable weights within UWMMSE are parameterized using graph neural networks (GNNs), where the time-varying underlying graphs are given by the fading interference coefficients in the wireless network. These GNNs are trained through a gradient descent approach based on multiple instances of the power allocation problem. We show that the proposed architecture is permutation equivariant, thus facilitating generalizability across network topologies. Comprehensive numerical experiments illustrate the performance attained by UWMMSE along with its robustness to hyper-parameter selection and generalizability to unseen scenarios such as different network densities and network sizes.