The Research Of DOA Estimation Algorithm For MIMO Radars

The concept of MIMO radar is proposed due to the inspiration of the mature MIMO technology in communication sector, which uses antenna arrays both in the transmitter and receiver, and transmits mutual orthogonal signal. Due to the applications of diversity techniques, MIMO radar can improve the performance in direction of arrival (DOA) estimation and target detection.This paper concentrates on the DOA estimation problem in MIMO radar system. Firstly, this paper introduces signal model of MIMO radar as well as the signal processing, and then a comparison is made between MIMO radar and the phased array radar. Secondly, this paper introduces four typical algorithms for DOA estimation: Capon algorithm, MUSIC algorithm, ESPRIT algorithm and Maximum Likelihood algorithm. The simulation shows that the MIMO radar's performance in DOA estimation is much better than single port array signal processing radar system, and the performance is improved with the increased inter-element spacing. Then, we studied the use of MUSIC method in MIMO radar farther, and a new way of estimating covariance matrix is proposed, the principle of this method is discussed in this paper. Theoretical analysis shows that the estimation performance of the new method is better than the traditional method. The conclusion is testified correctly by formula derivation and computer simulation. Finally, the new way of estimate covariance matrix is used in the M-SSMUSIC algorithm. The computer simulation shows that the proposed method improves the performance of MIMO radar DOA estimation compared to both SSMUSIC algorithm and M-SSMUSIC algorithm.