Adaptive Beamforming Technique For Large-scale Arrays Based On Sparse Array Sampling

Large-scale array beamforming with high degrees of freedom(DOF)has become more and more important recently.And the adaptive beamforming based on sparse array sampling can reduce the expenditure of hardware and obtain the same number of DOF with the large-scale array.In this paper,we stuty the adaptive beamforming technique for linear large-scale arrays with sparse array sampling and robust methods,and then we extend the structure of the sparse array to two dimensions.Firstly,the adaptive beamforming techhnique is studied.The signal model of one dimensional arrays and two dimensional arrays are introduced,and the traditional minimum variance distortionless response(MVDR)technique is derived.From the simulation results,we can find the performance of the MVDR beamformer is influenced by the error of system,then several well-known robust methods are introduced against the system error.This part provides the theoretical support for the application of the MVDR beamformer and robust method on the beamforming system based on sparse array sampling.Secondly,the adaptive beamforming technique for one dimensional large-scale array based on sparse array sampling is studied.During this part,the samlipng model of nested array and coprime array is built,and the augmented covariance matrix is constructed using the second-order statistics of the sparse array.Then the weight vector with the same dimension with the large-scale array can be obtained by using the reconstructed augmented covariance matrix.The beamformer based on sparse array sampling reduces the cost of hardware and gets the high DOF effectively.Thirdly,a noval robust beamforming method for the system of sparse array sampling is proposed.The steering vector of the desired signal can be estimated using the augmented covariance matrix of sparse array with imprecise prior information of array geometry and the range of direction of arrival(DOA)of desired signal.Then the desired signal in the sample data can be removed by reconstructing the covariane matrix of interferences plus noise.This method performs well against the steering vector error and the covariance matrix mismatch caused by finite sample size.Finally,the beamforming system for two dimensional large-scale array based on two dimensional sparse array is proposed.The array geometries of two dimensional nested array and coprime array and their DOF are analyzed,and some suggestions for two dimensional sparse array are given.Then the methods of constructing augmented covariance matrix and calculating are derived.The beamforming system based on two dimensional sparse array sampling can obtain the same DOF with the two dimensional large-scale array.