Study On Robust Adaptive Beamforming Algorithm In Complex Perturbated Environment

With the rapid development of global mobile communication, the lack of spectrum resources is more and more serious. In order to accommodate more mobile users in the limited spectrum resources, smart antennas technology is implemented which develop for half a century. Adaptive beamforming becomes one of the key technologies of smart antenna, it has been widely used in the fileds of radar, communications, sonar, navigation, radio astronomy and biomedical engineering and so on. But, in practical communication environment, the performances of adaptive beamforming algorithms are known to degrade severely in the presence of even slight mismatches as a consequence of look direction and signal pointing errors.Therefore, it has important theoretical and practical value to improve robustness and reduce the complexity cost.In this thesis, we introduce five basic performance criteria and six basic algorithms about adaptive beamforming, emphasize to study robust adaptive beamforming, analyze and compare the advantages and disadvantages of various algorithms. In the real perturbated environment, to accourt for mismatches, the improved robustness LSMI algorithm is proposed to improve the perfprmance of traditional LSMI algorithm. The proposed algorithm implements stably in more complex communication environment, and provides fast convergence rate, and achieve better results in the small training sample size.In the complex perturbated environment, the performance of traditional beamforming algorithm dramatically degrade, so people have proposed many effective robust beamforming algorithms. Diagonal loading is a simple and effective algorithm to improve the robustness of robust adaptive beamforming which use diagonal loading value to modify array covariance matrix. But the classical LSMI algorithm have the fixed diagonal loading value, which is determined only by experience, so far,there is no good way to solve this problem. To overcome this shortcoming, this paper presents an improved robust LSMI algorithm, which use the value of variable diagonal loading to amend covariance matrix,in situations of small training sample size. In order to reduce complexity cost, the steepest descent method is used to update the weight vector. Simulation results show that the algorithm is suitable for the small size of samples, and can reduce the complexity cost, provide robustness against mismatches, improve output SINR, and implement in situation of small training sample size.