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.
Original language | English |
---|---|
Article number | 9403959 |
Pages (from-to) | 6004-6017 |
Number of pages | 14 |
Journal | IEEE Transactions on Wireless Communications |
Volume | 20 |
Issue number | 9 |
DOIs | |
State | Published - Sep 2021 |
Externally published | Yes |
Funding
Manuscript received September 4, 2020; revised January 15, 2021; accepted March 30, 2021. Date of publication April 13, 2021; date of current version September 10, 2021. Research was sponsored by the Army Research Office and was accomplished under Cooperative Agreement Number W911NF-19-2-0269. The associate editor coordinating the review of this article and approving it for publication was J. Hoydis. (Corresponding author: Santiago Segarra.) Arindam Chowdhury and Santiago Segarra are with the Department of Electrical and Computer Engineering (ECE), Rice University, Houston, TX 77005 USA (e-mail: [email protected]; [email protected]).
Keywords
- algorithm unfolding
- deep learning
- graph neural networks
- power allocation
- WMMSE