Research On Adaptive Beamforming Of Towed Array With Large-aperture

Antisubmarine warfare is a crucial underpinning for future navy missions. For a long time, the problem of detecting increasingly quiet submarines has been remaining a very active research area in underwater acoustics. Towed line array sonar has become the important equipment in antisubmarine warfare by its advantage of configuration and characteristics, whereas beamforming plays an important role. However, towed array always leads to problems of shape distortion and snapshot-deficient because of flexibility and large-aperture, all of which influence the whole performance available in sonar systems. In these thesis, the problems below have been systematically analyzed and the corresponding methods to resolve also been studied. The main contents are as follows:1. Towed array shape estimation techniques are studied thoroughly. Based on theoretical analysis, the current algorithms are comprehended; the importance of shape estimation has also been verified by means of computer simulation for a dynamic long towed array.2. The reasons of snapshot-deficient are analyzed thoroughly. Three robust beamforming techniques are introduced, including diagonal loading, null broadening and eigenspace-based beamformer. In order to alleviate the performance degradation of adaptive beamfonning methods in presence of fast-moving strong interferers, a robust adaptive beamforming algorithm is presented. Null broadening technique is used to suppress interferers, and diagonal loading is employed to enhance its robustness meantime. The paper also provides the rule how to make a choice between eigenspace-based beamformer and SMI beamformer.3. A new sub-array beamspace adaptive beamforming algorithm is proposed. The correlation between source signals is equivalently increased with finite snapshots at array output. This will cause the performance degradation for existing sub-array adaptive beamforming algorithms. The new algorithm decreases the correlation between source signals by averaging the multiple-beam-output covariance matrix of each sub-array, and thereby improves the performance of adaptive beamforming. Computer simulations showed that the proposed algorithm had higher output signal-to-interference and noise ratio (SINR) with fewer snapshots. In addition to the improved SINR, the new algorithm also suppresses the noise received by array elements efficiently. The diagonal loading and eigenspace-based technique are extended into sub-array beamspace to improve robustness. Computer simulation results show that sub-array beamspace algorithm retains the super performance of its element space counterpart.