Deep Learning Beam Optimization in Millimeter-Wave Communication Systems

TL;DR

This paper introduces a combined neural network and fixed point algorithm approach to optimize beam configurations, user assignments, and power allocation in millimeter-wave systems, enhancing fairness and reducing complexity.

Contribution

It presents a novel one-shot neural network prediction method for beam configurations integrated with fixed point algorithms for joint resource optimization in millimeter-wave communications.

Findings

Reduces beam search complexity significantly.

Ensures optimal power and user assignment given predicted beams.

Achieves fair rate allocation among users.

Abstract

We propose a method that combines fixed point algorithms with a neural network to optimize jointly discrete and continuous variables in millimeter-wave communication systems, so that the users' rates are allocated fairly in a well-defined sense. In more detail, the discrete variables include user-access point assignments and the beam configurations, while the continuous variables refer to the power allocation. The beam configuration is predicted from user-related information using a neural network. Given the predicted beam configuration, a fixed point algorithm allocates power and assigns users to access points so that the users achieve the maximum fraction of their interference-free rates. The proposed method predicts the beam configuration in a "one-shot" manner, which significantly reduces the complexity of the beam search procedure. Moreover, even if the predicted beam configurations are not optimal, the fixed point algorithm still provides the optimal power allocation and user-access point assignments for the given beam configuration.