Learning Energy-Efficient Modular Arrays under Hardware Non-linearities

2026-05-21

AI Summary

This work addresses the energy efficiency optimization of extremely large-scale sparse aperture arrays under power amplifier (PA) nonlinearities by proposing an unsupervised deep neural network (DNN) approach that jointly optimizes power allocation, total transmit power scaling, and modular subarray activation. To the best of our knowledge, this is the first application of unsupervised DNNs to joint energy efficiency optimization in the presence of nonlinear distortion. The method leverages Bussgang decomposition to model PA nonlinearities and integrates singular value and geometric channel features to enable end-to-end decision-making. Experimental results demonstrate that the proposed scheme significantly improves energy efficiency compared to conventional sparse array strategies while maintaining spectral efficiency.

Abstract

This paper investigates the joint optimization of power allocation and antenna activation in sparse extremely large aperture array systems operating under power amplifier non-linearities. We first derive an analytical expression for the achievable spectral efficiency (SE) of point-to-point MIMO channels affected by non-linear distortions using the Bussgang decomposition. To address the combinatorial and non-convex nature of the energy-efficiency (EE) maximization problem, we employ an unsupervised deep neural network (DNN) that learns the non-linear mapping between the channel state information and the optimal EE operating point. The DNN jointly predicts distortion-aware power allocation, total transmit power scaling, and modular sub-array activation based on singular-value and geometric channel features. Numerical results demonstrate that the proposed DNN-based arrays achieve significant EE gains over the conventional sparse arrays.

Problem

• energy efficiency
• modular arrays
• hardware non-linearities
• antenna activation
• power allocation

Innovation

• energy efficiency
• non-linear distortions
• modular arrays
• unsupervised DNN
• antenna activation