A Research Of Robust Anti-jamming Algorithm Of Digital Array Radar

Digital beamforming techniques is one of the key techniques of array signal processing, and it is usually used as the anti-jamming algorithm in the digital array radar. In practical application, the steering vector mismatch and the presence of the desired signal with high power often cause a serious recession of the beamforming performance. Therefore, robust adaptive beamforming algorithm has become the focus of related field. The main research content of this thesis is analyzing the performance of MVDR algorithm for the finite sample data when there is a mismatch in the steering vector of the desired signal. Moreover, this thesis has studied some algorithms based on the steering vector estimation and the reconstruction of interference plus noise covariance matrix. The specific work of this thesis is presented as follows:.Firstly, the basic model of array signal processing, the foundation of adaptive beamforming algorithm, is introduced. Normalized beam pattern and the output SNR(signal-to-noise ratio) is given as the performance measures for beamforming algorithm. Then several common robust beamforming algorithms are presented and their performance are compared through the simulation experiments.Next, the influence of desired signal power on the performance of the MVDR algorithm when there is a mismatch of desired signal steering vector is analyzed in detail. By substituting the steering vector mismatch into the weight vectors of MVDR algorithm, a novel expression of the weight vector can be obtained for the finite sample data. In addition, based on this new weight vector, the performance of MVDR algorithm when there is an error of signal steering vector is analyzed in detail. To verify the correctness of the theoretical analysis, the results of theoretical analysis and simulation results are compared through the simulation experiments.Then, several adaptive beamforming algorithm based on steering vector estimation are studied. These methods estimate the real steering vector by using a priori information, which gives out a range of the arrival direction of the desired signal. Simulation results show that these methods have robustness to some extent. But all of them are restricted by the desired signal power. Some of them only suit for the situation when desired signal is not strong, and others just fit for the situation when desired signal power is stronger than interference signal power.Finally, two kinds of robust adaptive beamforming are studied. These beamformers is able to break the restriction of desired signal power and is based on interference and noise covariance matrix reconstruction. Moreover, a new robust beamforming method by using the reconstructed covariance matrix via sparse reconstruction algorithm and the estimation of steering vector to compute the weight vector is proposed. The good performance of this method is verified by simulation experiments when the mismatch of steering vector and the strong power of desired signal appear together.