Detection of Multiple Closely Spaced Targets in Low SNR Conditions using MUSIC Algorithm

Abstract

Radar is an electromagnetic system used for detecting, locating and tracking targets. To enhace the multi target detection capability of digital receivers in Electronic Warfare applications, high resolution algorithms are required. Multiple signal classification (MUSIC) is one of such prominent algorithm, but its performance is degraded in the presence of noise and when the targets are in close proximity.Therefore, in this paper MUSIC algorithm is modified at two stages.In the first stage Savitzky Golay filter is used as Pre-processing filter and at next stage noise removal technique is used for estimating the target eigenvalues and frequencies. The results showed that proposed algorithm is capable of distinguishing 3 targets with Signal to Noise Ratio (SNR) of upto − 12 dB and 4 targets with SNR of -9dB with good range resolution compared to available literature. Optimized computation complexity of the proposed method is making it as superior compared to other methods.