Research On DOA Estimation Using Acoustic Vector Sensors Array In Nonuniform And Correlated Noise

Underwater acoustic array signal processing has the advantages of strong anti-interference ability and high array gain.It has developed a lot of research results.The target DOA estimation algorithm is a basic problem in underwater acoustic array signal processing.In addition,noise has always been the main factor affecting the accuracy of the DOA estimation algorithm.The predecessors often assumed that the noise model was a white noise model.However,in engineering practice,ideal white noise does not exist,which will cause severe performance degradation or even failure of the algorithm.A vector sensor can measure the three orthogonal components of sound pressure and sound particle velocity at a point in space at a common point.Compared with the traditional sound pressure sensor,a single vector sensor or its array has more advantages,so it has become a research trending of scholars.Based on the advantages of vector sensors,this paper conducts research on target orientation estimation based on acoustic vector sensors.At the same time,it studies the target angle estimation method in the case of non-uniform correlated noise.In an isotropic noise field,the noise correlation is caused by the array element spacing.When the array element spacing is less than half a wavelength,the noise between the array elements has a clear correlation.At this time,the background noise field is a non-uniform correlated noise field.The first part of this paper studies a robust maximum likelihood DOA estimation method to resist the effects of non-uniform correlated noise.Simulation results show that under the condition of non-uniform correlated noise,the performance of the robust maximum likelihood DOA estimation algorithm is significantly better than the traditional maximum likelihood DOA estimation method and the traditional MUSIC method.The second part verifies the performance of conventional beamforming,Generalized Sidelobe Cancellation algorithms and robust Generalized Sidelobe Cancellation algorithms under non-uniform correlated noise conditions.The results show that,under normal circumstances,the Generalized Sidelobe Cancellation algorithm is more suitable for azimuth estimation research than the conventional beamforming algorithm and the robust Generalized Sidelobe Cancellation algorithm.On the background of non-uniform correlated noise with low signal-to-noise ratio,the robust Generalized Sidelobe Cancellation algorithm is more resistant to strong interference from the background than the conventional beamforming algorithm,and the energy of the main lobe is higher than that of the Generalized Sidelobe Cancellation algorithm.In addition,the robust Generalized Sidelobe Cancellation algorithm is robust to array amplitude and phase errors.The third part of this paper uses pool test data to verify the robust maximum likelihood estimation algorithm and the robust Generalized Sidelobe Cancellation algorithm.The processing results further verify the advantages of the robust maximum likelihood estimation algorithm and the robust Generalized Sidelobe Cancellation algorithm.