Multi-User Continuous-Aperture Array Communications: How to Learn Current Distribution?

TL;DR

This paper introduces L-CAPA, a deep learning framework that learns current distribution policies for continuous aperture array systems, enabling near-optimal multi-user communication performance with low complexity.

Contribution

The paper proposes a novel deep learning approach using graph neural networks to learn current distributions in CAPA systems, addressing non-convex optimization challenges without closed-form solutions.

Findings

L-CAPA achieves performance close to the theoretical upper bound as antenna count increases.

The framework effectively learns current distributions with low inference complexity.

Simulation results validate the effectiveness of the proposed method.

Abstract

The continuous aperture array (CAPA) can provide higher degree-of-freedom and spatial resolution than the spatially discrete array (SDPA), where optimizing multi-user current distributions in CAPA systems is crucial but challenging. The challenge arises from solving non-convex functional optimization problems without closed-form objective functions and constraints. In this paper, we propose a deep learning framework called L-CAPA to learn current distribution policies. In the framework, we find finite-dimensional representations of channel functions and current distributions, allowing them to be inputted into and outputted from a deep neural network (DNN) for learning the policy. To address the issue that the integrals in the loss function without closed-form expressions hinder training the DNN in an unsupervised manner, we propose to design another two DNNs for learning the integrals. The DNNs are designed as graph neural networks to incorporate with the permutation properties of the mappings to be learned, thereby improving learning performance. Simulation results show that L-CAPA can achieve the performance upper-bound of optimizing precoding in the SDPA system as the number of antennas approaches infinity, and it is with low inference complexity.