Research On High Resolution Techniques For Direction Finding Of Underwater Target

High-resolution Direction of Arrival (DOA) estimation is always a hot research area in array signal processing. The breakthrough of this technique is greatly meaningful in many research fields such as radar, sonar, communication, national defense and so on. Especially the underwater acoustics high-resolution multi-target direction finding technique is the key to underwater searching and ocean exploration. Aiming at application in engineering, the high-resolution direction-finding theory, algorithms, and the calculation procedures are studied in this paper. Main research results and achievements are summarized as follows:1. The classical subspace high-resolution DOA estimation methods are disserted. The theory for the MUSIC, MNM, TLS-ESPRIT and GEESE algorithms and the calculation procedures are discussed. The integrated performance evaluation indexes are presented. According to this indexes, the performance of the aforementioned subspace methods are analyzed by computer simulation. Based on the analysis, the practicability of methods is evaluated.2. The methods for high-resolution DOA estimation based on multi-sensor SPMP are studied, which computation complexity is simple and performance is unaffected by unknown correlated noise. One is based on energy criterion (Also called E-SPMP); another improved on E-SPMP is based on matrix-norm criterion (Also called N-SPMP). On the basis of theoretical study, larger amount of computer simulation and statistical analysis are conducted to investigate the performance of the two methods. It is shown that N-SPMP algorithm reduces largely the RMSE of estimation compared with E-SPMP and MUSIC algorithm at low SNR and its performance approximates MUSIC algorithm at high SNR, but the computation simplicity is still kept. The computer simulation results are also shown the estimation performance of the N-SPMP and E-SPMP algorithms is unrelated to the space interval of the signals in a certain extent. So the N-SPMP algorithm can divide signals that has small space interval at low SNR. Otherwise, the N-SPMP algorithm has low RMSE still in small array. Specially, the N-SPMP algorithm still has good performance under unknown correlated noise.