The Research On Underwater-object-detect Array And DOA Estimation

Underwater-object-detect array and array signal processing play important roles in underwater security defend. Both array designing and array signal processing with high-resolution are the most important directions in underwater acoustics. Direction level of sensors is ignored when the array works under low frequency. Because the wavelength is long enough, we take no account of dispersion. However, if the frequency is much higher, the direction level of sensors and dispersion must be considered. The paper put forward the necessity of research on underwater-object-detect array first, and then discussed the relationship between array region and CRB in condition of sensor direction level. We did some research on DOA estimation in the case of coherent signals and sparse circular array. It is proved that the algorithm performs well under the condition of sparse circular array. The main work of the paper is summarized as follows:(1) The relationship between array region and CRB was discussed on the base of mode space method. And then the relationship between sensor direction level and CRB was analyzed. Afterwards we put forward the main method of array design.(2) The basic theory of array signal processing was introduced, and then we explained the reason why MUSIC failed when there were coherent signals. After this we put forward shift reference sensor method, which not only has high-resolution on coherent signals, but also keeps the effective aperture of array.(3) The influence of dispersion of a rigid cylindrical baffle on high-resolution direction finding was discussed. Then the new algorithm was introduced, it can make a high-resolution estimation under sparse circular array. It is much more useful than CBF under some conditions.(4) The experimental parameters were chosen, such as the region of array, the number of sensor, etc. Besides, quadrature demodulator was designed and software for the system was developed. After this, we estimated the azimuth angle of target using the algorithm put forward above. It is proved that the algorithm is effective and accurate in the case of sparse circular array.