A Sequential Estimation Approach for Performance Improvement of Eigenstructure-Based Methods in Array Processing

Abstract

In this correspondence, we present a simple and noniter- ative approach that lowers noticeably the threshold signal-to-noise ra- tio (SNR) of the eigenstructure-based techniques for estimating the di- rections of arrival (DOA's) of multiple narrow-band sources in passive sensor arrays. This approach removes effectively the spatial interfer- ences among sources through the sequential estimation of the DOA's. We develop a sequential multiple signal classification (MUSIC) algo- rithm by applying the approach to the MUSIC algorithm, although it can be equally well applicable to other eigenstructure-based methods. We also present a recursive computational procedure (RCP) that re- duces significantly the computational complexity of the proposed al- gorithm by transforming the computation of Hermitian forms into that of only inner products of vectors. Computer simulation results that demonstrate the resolution performance of the proposed algorithm are included. Since this algorithm is simple and noniterative, and also pro- vides high resolution, it may be used on its own, or be used to provide good initial estimates for more complex iterative algorithms.