Research And Simulation On Adaptive Beamforming Algorithms Of Smart Antenna

With the rapid increase in mobile users,the conflict between the increasing of Mobile traffic and limited spectrum utilization is more and more seriously.Because the smart antenna technology has a unique spatial-filtering characteristic,space division multiplexing has been achieved.It makes channel capacity Increase and uses spectrum resources to the greatest extent.Therefore,smart antenna technology has become one of the hotspots in a wireless communication technology.It is formed by some antenna arrays.Smart antenna uses modern digital processing techniques and appropriate adaptive algorithms to form space direction beam dynamically.In order to make full use of the desired signals,we make the main beam toward the direction of arrival and make the null point toward the interfering signals.Adaptive beamforming algorithm is a crucial problem of the smart antennas.First,the thesis introduces basic principles and working modes of smart antenna technology.Four common optimal criterion are analyzed,which use in adaptive beamforming algorithms.Second,combined with sudden signal characteristics,some known portion unknown sequence features for the user,detailed study,an improved algorithm based on the traditional CMA and LS-CMA algorithm is presented to solve problems that we know some signals and do not known others.The improved algorithm obtains signal subspace and noise subspace by the way of decomposition of feature space.The initial weights of the beam tracking are obtained by the cancellation of the interference signal.The dynamic least square constant modulus algorithm is used to track the expected signal.The simulation results show that the improved algorithm has fast convergence speed,stable performance and good effect on the suppression of interference.Third,for smart antenna DOA estimation algorithms,if we have coherent sources and incident angle signal interval is small,the thesis presents an improved algorithm on the basis of classical MUSIC algorithm and the modified MUSIC algorithm.Using the correlation matrix information,received signals form a new covariance matrix,which can improve the utilization of data.Then singular value decomposition is used to calculate spatial spectrum.Simulation results show that the improved algorithm has better capability of measuring direction in low signal noise ratio and low number of snapshots situations.At last but not least,this thesis concludes the research work and looks forward to solving problems in the research process and future work.