Effects of hydrophone gain errors and self-noise on the DOA performance

This paper studies the effects of hydrophone gain errors and self-noise on the direction-of-arrival (DOA) performance of three methods: the conventional beamforming (CBF) method, the minimum variance distortionless response (MVDR) method and the multiple signal classification (MUSIC) method. Simulations for a 16-hydrophone uniform circular array whose diameter is 2 m are provided in three different scenarios, in which the incident signal is supposed to be a narrow-band signal. The results show that both the gain errors and the self-noise make the main-lobes of the beam patterns wide and the side-lobes high, and the self-noise caused by data channels has less influence on the DOA performance than the gain errors caused by the external pressure. Experiments with a 16-hydrophone uniform circular array at a depth of 3500 m in the South China Sea show that the three methods cannot obtain the direction of a real target at a low signal to noise ratio (SNR) when two hydrophones have large self-noise. By contrast, when the amplitude of each channel signal is normalized individually, the direction of the real target can be estimated using the above-mentioned three methods. If gain errors of some hydrophones are large, amplitude normalization can eliminate the errors very well. Consequently, amplitude normalization can reduce the negative effects of hydrophone gain errors and self-noise and improve the DOA performance.