A Machine Learning Approach to DoA Estimation and Model Order Selection for Antenna Arrays with Subarray Sampling

TL;DR

This paper introduces neural network-based methods for direction of arrival estimation and model order selection in antenna array systems with subarray sampling, achieving improved accuracy and lower computational complexity especially with limited data.

Contribution

It presents novel neural network schemes that outperform existing methods in DoA estimation and model order selection for subarray sampled antenna arrays.

Findings

Outperforms existing estimators in mean squared error.

Achieves higher model selection accuracy.

Operates efficiently with low snapshot data.

Abstract

In this paper, we study the problem of direction of arrival estimation and model order selection for systems employing subarray sampling. Thereby, we focus on scenarios, where the number of active sources is not smaller than the number of simultaneously sampled antenna elements. For this purpose, we propose new schemes based on neural networks and estimators that combine neural networks with gradient steps on the likelihood function. These methods are able to outperform existing estimators in terms of mean squared error and model selection accuracy, especially in the low snapshot domain, at a drastically lower computational complexity.