A Supervised Learning Framework for Joint Angle-of-Arrival and Source Number Estimation

TL;DR

This paper introduces a machine learning framework that jointly estimates the number of sources and their angles of arrival, effectively handling variable source counts and outperforming traditional methods under various SNR conditions.

Contribution

It presents a novel supervised learning approach using classifiers to jointly estimate source number and AOAs, accommodating variable source scenarios unlike prior fixed-source methods.

Findings

The proposed estimator outperforms MUSIC at certain SNRs and resolutions.

Performance varies with SNR and FOV resolution, with lower resolution better at low SNR.

The method effectively estimates source number and AOAs jointly.

Abstract

Machine learning is a promising technique for angle-of-arrival (AOA) estimation of waves impinging a sensor array. However, the majority of the methods proposed so far only consider a known, fixed number of impinging waves, i.e., a fixed source number. This paper proposes a machine-learning-based estimator designed for the case when the source number is variable and hence unknown a priori. The proposed estimator comprises a framework of single-label classifiers. Each classifier predicts if waves are present within certain randomly selected segments of the array's field of view (FOV), resulting from discretising the FOV with a certain (FOV) resolution. The classifiers' predictions are combined into a probabilistic angle spectrum, whereupon the source number and the AOAs are estimated jointly by applying a probability threshold whose optimal level is learned from data. The estimator's performance is assessed using a new performance metric: the joint AOA estimation success rate. Numerical simulations show that for low SNR (-10 dB), a low FOV resolution (2$^\circ$) yields a higher success rate than a high resolution (1$^\circ$), whereas the opposite applies for mid (0 dB) and high (10 dB) SNRs. In nearly all simulations, except one at low SNR and a high FOV resolution, the proposed estimator outperforms the MUSIC algorithm if the maximum allowed AOA estimation error is approximately equal to (or larger than) the FOV resolution.