Research On Robust Adaptive Beamforming Based On Analysis Of Array Error

Adaptive beamforming is able to adjust the weight adaptively, make the mainlobe aimat the desired signal and make the interferences signal nulling, which could suppress theinterferences signal effectively and strengthen the desired signal. However, in the practicalapplication, the performance of conventional adaptive beamforming methods is known todegrade severely in presence of signal model mismatches. There are channel amplitude andphase error, element location error, beam point error and low snapshot number error etc,which could cause signal model mismatches. Therefore, robust adaptive beamformingalgorithm and its implementation is one of the hot research topics in the field of array signalprocessing currently.The performance of the conventional adaptive beamforming affected by different erroris analyzed based on the introduction of adaptive array signal mode1and opticalbeamforming criterion. According to the above analysis, when the desired signal is presentin the training snapshots, the errors will result in sidelobe level elevation, mainlobedistortion and output signal-to-interference-plus-noise ratio (SINR) reduction. In this case,the performance of conventional adaptive beamforming degrades severely due to thedesired signal cancellation effect. However, when the desired signal is not present in thetraining snapshots, adaptive beamforming algorithm is not sensitive to the errors androbustness is strong.Some classical robust adaptive beamforming algorithms are elaborated, meanwhile,and their advantages and disadvantages are analyzed. In order to solve the shortage of theexisting adaptive beamforming algorithms, a robust adaptive beamforming for actualsystem is proposed based on the reconstruction of interference covariance matrix andmismatched steering vector compensation. In the proposed method, the interferencecovariance matrix is firstly reconstructed by using Root-MUSIC method to estimateDirection-of-Arrival (DOA) of signals, where the desired signal component is removedfrom the training snapshots. Subsequently, the mismatched desired signal steering vector iscompensated by solving a quadratically constrained quadratic programming problem.Simulation results show that the performance of proposed algorithm outperforms theexisting robust adaptive beamforming, and the output SINR is close to optical value.Therefore, the proposed algorithm could be significantly effective for actual system.When the desired signal is present in the training snapshots, the performance of conventional orthogonal projection (OP) adaptive beamforming degrades severely due tothe desired signal cancellation effect. To overcome this deficiency, the improvedorthogonal projection adaptive beamforming by using reconstructed interference and noisecovariance matrix is proposed. In the proposed algorithm, the interference and noisecovariance matrix is firstly reconstructed by integrating the Capon spatial spectrum over aregion separated from the desired signal direction. Subsequently, the interference subspaceis estimated, and then the adaptive weight vector is calculated using conventional OPalgorithm. The simulation results show that the corresponding output SINR performance ofproposed algorithm is almost same with the optimum beamforming with the desired signalin the training snapshots. Therefore, the proposed algorithm is significantly effective for theactual system. When the OP algorithm is applied in subarray level processing, theperformance of OP algorithm will be degraded. Therefore, the orthogonal projection atsubarray level based on normalization is proposed. According to simulate results andanalysis, it is shown that the performance of the proposed algorithm is close to the optimalalgorithm.Finally, the full article is summarized, and the further research of robust adaptivebeamforming is provided.