Research On Widely Linear Filter Methods Of Underwater Acoustic Signals

The instantaneous frequency estimation of underwater acoustic pulse signals and the target resolution problem under multi-target interference conditions have always been difficult and hot issues in the field of underwater acoustic signal processing.In recent years,the rapid development of widely linear filtering technique dealing with non-circular signals has become an emerging research direction in the field of signal processing.Compared with the linear processing method,the wide linear processing method considers the complex signal and its conjugate component simultaneously,which brings the result that the amount of information has been enlarged and the performance of signal processing has been improved.The sound pressure and particle velocity response of the differential pressure vector hydrophone share an orthogonal characteristic,which provides the possibility to utilize a wide linear filtering method.In this paper,based on the research of the basic theory of wide linear time domain filtering and spatial filtering technology,with the method of joint processing of the sound pressure and particle velocity response of the vector hydrophone,the researches focus underwater acoustic widely linear instantaneous frequency estimation method and underwater acoustic array signal widely linear beamforming method.The main contributions and innovations of this paper are summarized as follows:(1)A joint widely linear processing method of the sound pressure and particle velocity response of a differential pressure vector hydrophone is proposed.A complex signal composed of sound pressure and the particle velocity response of the single vector hydrophone for joint processing has been proposed.The widely linear time domain filtering method used for instantaneous frequency estimation would improve the frequency estimation accuracy.The widely linear beamforming method is used in order to improve the multi-target resolution of the vector hydrophone array.(2)A widely linear instantaneous frequency estimation algorithm based on ACLMS algorithm for single vector hydrophone is proposed.In order to improve the estimation accuracy of instantaneous frequency of underwater acoustic signals,a complex signal composed of the sound pressure and particle velocity response of vector hydrophones as the real part and imaginary part respectively was constructed.Three widely linear instantaneous frequency estimation equations based on ACLMS algorithm are derived.Through simulation,anechoic tank experiments and sea test data analysis,the algorithm proposed in this paper was validated.The results show that the proposed algorithm performs better than the traditional short-time Fourier transform method and the CLMS algorithm based on linear filtering model under high SNR conditions.(3)The widely linear beamforming algorithms based on vector hydrophone array signals are studied.In order to improve the underwater acoustic array signal processing performance and the multi-target resolution,the widely linear beamforming methods are applied to the processing of the vector hydrophone array signals,by using the sound pressure signals and particle velocity response signals of the vector hydrophone array as the desired signal.Through simulation analysis,the performance of widely linear beamforming and traditional linear MVDR beamformer are compared.The results show that the widely linear beamforming method improves the multi-target resolution and output signal-to-noise ratio.The feasibility of the algorithm are also validated by the sea trial data processing analysis.