Research On Robust Adaptive Beamforming And Forming Technology

Beamforming and shaping technology are important components in the field of array signal processing.It obtains the main lobe of the desired signal direction based on the array weight vector under certain criteria,and forms a null trap in the interference direction,which has achieved the purpose of spatial filtering.It is widely used in signal processing such as radar,satellite navigation,mobile communication,medical diagnosis,voice and sonar.However,as the application background of beamforming technology becomes more and more complex,the beamforming effect is getting worse due to inaccurate estimation of the sample covariance matrix.For the beamforming technology,the existing shaping algorithm has a problem that the calculation amount is too large and the convergence speed is slow,so that it is difficult to apply in real time.Therefore,it is of great theoretical and practical value to study robust beamforming algorithms and efficient beamforming algorithms in non-ideal environments.The main work and innovations of this thesis are summarized as follows:This thesis first introduces the mathematical model and basic parameters of the adaptive beamforming algorithm.At the same time,it introduces several common optimal beamforming criteria,and analyzes the effects of array covariance matrix error and steering vector mismatch beamformer.Algorithm research laid the theoretical foundation.Secondly,for the problem of performance degradation of adaptive beamforming algorithm in finite sampling data,a robust beamforming algorithm under the characteristic interference canceller is proposed.The algorithm suppresses the influence of the noise subspace by appending the Dolph-Chebyshev taper weight to the optimal weight vector of the traditional beamformer and constraining the array weight vector to the noise subspace.In order to overcome the influence of array steering vector error on the performance of beamformer,a robust adaptive beamforming algorithm based on orthogonal decomposition of mismatch error is proposed.The algorithm uses the orthogonal component of the mismatch error to correct the desired signal steering vector.By combining the two algorithms,a hybrid robust algorithm which can reduce the influence of the covariance matrix on the beamformer at low sampling and overcome the expected signal steering vector mismatch is obtained.Finally,in order to solve the commonly used beamforming algorithm,the computational complexity is large,the convergence speed is slow,the optimization parameter adjustment is complex,and it can not be directly applied to the online real-time calculation.An improved iterative minimum variance pattern synthesis method is proposed.The method uses the pattern sub-region weighted approximation and the pattern phase de-constraining technique,and uses the weight coefficient obtained in the previous time to calculate the maximum gain and pattern phase of the pattern,and as the maximum gain and pattern phase expected in the next iteration.