DoA Estimation Using Neural Network-Based Covariance Matrix Reconstruction

In this paper, we discuss a new approach to direction of arrival estimation for systems with subarray sampling. We propose to estimate the covariance matrix of the full array from the sample covariance matrices of the subarrays using a neural network. This technique enables the estimation of more sources than radio frequency chains by applying a MUSIC estimator to the reconstructed full covariance matrix. The proposed method is able to outperform classical estimators and has some benefits compared to a recently proposed machine learning-based technique for these systems, which models the direction of arrival estimation problem as a end-to-end regression task.