Research On Robust Adaptive Beamforming Based On Finite Snapshots And Lower SNR

Beamforming is a key technology in the array signal processing field. However,when sampling snapshots are finite and Signal Noise Ratio is lower, traditionalbeamforming algorithms will lead to the consequence which include beampatterndistortion and inaccurate direction of expected signal, as a result of poor output SignalInterfere-pulse-Noise Ratio, consequently more snapshots are needed to obtain goodspatial filtering effect. But more snapshots will result in high computational complexity,difficult hardware system implement, weak intercommunication instantaneity and soon. In order to solve such many problems of the traditional beamforming algorithms, arobust beamforming algorithm proposed in this dissertation is suitable for the conditionincludes finite sampling snapshots and lower SNR, and the main work include：Firstly, the principle of digital beamforming technology is briefly introduced, andsome typical beamforming algorithms’ simulations are given, and then some existingproblems of those algorithms, when finite sampling snapshots and lower SNR, areanalyzed.Secondly, Robust Capon Beamforming algorithm proposed by Jian-Li is introduced,its deducing process is so well given, and then the dissertation proves RCB algorithmin essence is a extension of diagonal loading algorithm, but its loading level is adjustedadaptively. Finally, MATLAB simulations of RCB algorithm and the fixed loadingdiagonal algorithm are given, which prove that RCB algorithm’s beampattern isundesirable when snapshots is finite and SNR is lower.Thirdly, to solve the problem that RCB algorithm’s performance degeneration whenfinite sampling snapshots and lower SNR, a novel algorithm combines conformalbeampattern with interfere rejection, is propose in the dissertation, and then itsdeducing process and computational complexity are analyzed. The algorithm estimatescovariance matrix with finite snapshots, and then the best weight vector by solve aoptimization problem is obtained. It not only enhances the algorithm’s robustness andreduce computational complexity, but also lower its spending time of iteration processand enhance its instantaneity. Finally, its efficiency and robustness with the aid ofexperiment simulation are proved.