Research On The Adaptive Beamforming Algorithm Of Smart Antenna

In recent years, with the continuous development of time-division multiplexing, frequency division multiplexing and code division multiplexing, the increasing of channel capacity has become very difficult. Nevertheless, depend on its unique characteristics of spatial filtering, smart antenna has made space division multiplexing real, and also promoted the channel capacity once again. As the core technology of smart antenna, beamforming regulates the weight of each array element in antenna array by the beamformer, so the output of the weighting network could achieve optimal according to the change of time-varying channel.Firstly, the basic concepts of smart antennas, the mathematical model and the adaptive beamforming technology are introduced, then some of the finest convergence criterions commonly used in the adaptive beamforming algorithm for smart antenna and the existing beamforming algorithm are also analyzed in this paper. Some improvements of most widely used variable step size least mean square (LMS) algorithm have been proposed in this paper. The step size of improved variable step size LMS algorithm decreases with the decrease of iterative error, and shows a relationship of hyperbolic tangent function with the error. At the same of ensuring the convergence speed and time-varying tracking capability, the new algorithm also can get a better steady state performance.Aim at the array error of the practical application, several common robust beamforming algorithms are analyzed and discussed, and one of the characteristics of sub-space (ESB) algorithm has been improved. The improved ESB algorithm judges the number of source based on the criteria of characteristic difference when coherent signals exist, and constructs the signal subspace and noise subspace according to the correct number of sources. Simulation results have showed the advantages and effectiveness of the proposed algorithm in coherent environment and low SNR environment. It overcomes the shortcoming of the ESB algorithm that cannot be used in coherent environment and low SNR environment, and make the ESB algorithm obtain a wider range of applications.