MIMO Sonar Transmit Energy Optimum Design for Multi-target Direction of Arrival Estimation

Abstract

In this paper, we propose a multiple-input multiple-output (MIMO) sonar operating in an environment with two or more located targets. A method for transmit energy optimum is proposed. The new technique enjoys the advantages of the MIMO sonar without suffering from the loss of phased-array sonar coherent processing gain, reduces the computational complex from the high dimensional matrix. The essence of the proposed method is to partition the transmit array into two subarrays. The rotational invariance property is established and the energy is evenly distributed in the desired region. Moreover, the beamspace matrix reduces the computational complexity significantly. The covariance matrix of received signals maintains stable irrelevance. Therefore, the subspace decomposition-based techniques such as Multiple Signal Classification (MUSIC) and the rotational invariance property techniques such as Estimating Signal Parameters via rotational invariance techniques (ESPRIT) can obtain better direction of arrival (DOA) estimation performance. Simulation results also show that the method can provide more accuracy DOA performance and better robustness in low signal-to-noise ratio (SNR) and limited snapshot number.