MIMO Sonar DOA Estimation Based on Improved Transmitting Diversity Smoothing (TDS)

Abstract

A multiple-input multiple-output (MIMO) sonar has an effect of transmission diversity smoothing (TDS), which is able to “de-correlate” the echoes of multiple targets automatically. Accordingly, a series of high-resolution DOA estimation methods can be adopted directly without using conventional spatial smoothing, and hence avoiding the losses of array aperture and degree of freedom. However, the performance of TDS-based DOA estimation methods would degrade seriously in low signal-to-noise ratio (SNR) condition. In this paper, to solve the problem we proposed a new MIMO sonar DOA estimation method based on improved TDS. In order to improve the SNR of data used for DOA estimation, two steps are used. Firstly, we use the sum of all transmitting signals to matched filter echoes at the receiving hydrophone array. And secondly, we truncate outputs of matched filters around the zero-delay point. Synchronously, the number of truncated snapshots is carefully considered to retain the TDS effect. After the two-step processing, the TDS effect is still available and SNR of the data to be processed is remarkably improved. And thus, the performance of the TDS-based DOA estimation method is effectively improved. Numerical simulations show that the proposed method could significantly improve the performance of TDS-based DOA estimation methods in low SNR environment.