Research On Low Frequency Robust Beamforming And Target Azimuth Estimation Of Vector Circular Array

In underwater acoustic propagation,the loss of signal propagation decreases with the decrease of frequency.In order to improve the detection distance of sonar equipment,low frequency is a major trend in the development of existing sonar equipment.Similarly,the detection of quiet underwater targets requires a higher performance of beamformer.In addition to vectorization and other means to improve the performance of sonar,the performance of sonar system is improved through some signal processing at the back end,such as the research on the robustness of array at low frequency is also a hot issue at present.In addition,due to the complexity of underwater battlefield conditions,the presence of coherent signals will degrade the performance of various high-resolution target azimuth estimation algorithms in the process of target detection.Therefore,this paper focuses on two aspects: one is the study of array robustness in low frequency;the other is the study of coherent signal decoherence.Firstly,this paper introduces the basic array signal processing algorithms,such as the circular array element domain receiving signal model,the anisotropic noise field model and CBF,MVDR,music,etc.of array signal processing.Then,it briefly introduces the definition of super gain,the difference and connection between array gain and directivity,and the basic theory of the commonly used dimension reduction to understand the coherent algorithm and the phase mode domain signal processing method.Next,this paper studies a low-frequency robust Super Gain Beamforming Method,which can determine the diagonal load value by setting different white noise gain loss values in the presence of sensor channel amplitude and phase error and array element position disturbance error.The change of diagonal load value can realize the continuous parametric adjustment between robustness and array gain,which is the contradiction of optimal The solution is the best compromise between the gain and robustness of the corresponding array.In addition,the decoherence processing method in the phase mode domain of the vector circular array is also studied.For narrow-band signals,the uniform circular array in the array domain is first transformed into a virtual uniform linear array in the phase mode domain through a series of processing and transformation,and then the spatial smoothing method isused to divide the sub array to recover the covariance matrix rank of the received data,so as to achieve the purpose of decoherence;for wide-band signals,the signal correlation matrix becomes nonsingular through the average in the frequency domain,this method It can effectively suppress background isotropic noise and estimate DOA of two coherent sources.Finally,the frequency domain average decoherence algorithm in phase mode domain is verified by the experiment of anechoic tank.