A Fast and Robust MUSIC Algorithm for Estimating Multiple Coherent Signals

Abstract

The conventional MUSIC algorithm demonstrates subpar estimation performance and unreliable results when confronted with the task of estimating coherent signals from multiple targets. Moreover, it suffers from high computational complexity and sluggish processing speed when applied to extensive datasets involving multiple sensors. In order to tackle these challenges, this paper presents an enhanced and expedient MUSIC algorithm for the estimation of multiple coherent signals. Drawing upon the ROOT-MUSIC algorithm based on the propagation operator, this algorithm introduces a spatial smoothing technique by substituting the original covariance matrix with the average of subarray covariance matrices. Through simulation results, it is demonstrated that the proposed algorithm not only resolves the issue of estimating multiple coherent signals but also achieves exceptional performance in terms of robustness and computational speed, even under low signal-to-noise ratio conditions.