Improving the Source Localization Performance of the Phase Array by Using the Radiation Pattern Error

In this paper, we analyze the effect of the radiation pattern error on the multiple signal classification (MUSIC) algorithm and demonstrate that the performance of the direction of arrival (DOA) estimation can be improved by utilizing the pattern error. We first analyze the performance of the MUSIC algorithm in the ideal phased array from two aspects of estimation accuracy and angular resolution. We derive the Cramer-Rao Lower Bound (CRLB) and the asymptotic formulas of the estimation’s error variance to quantitatively show the estimation accuracy. And the angular resolution is represented by the zero spectrum (ZS). We then research the variation of the estimation performance in the array with the pattern error and confirm the enhancement effect of the pattern error on the DOA estimation. Numerical results prove that the array with pattern errors has better DOA estimation performance than the ideal array, from CRLB, root mean square error (RMSE), probability of detection (Pd) and angular resolution.