The Research And Implementation On Algorithm In Array Antenna Adaptive Beam-Forming

The sensor array is widely used to receive and handle signals in the array signalprocessing which extracts the worthy characters and information composing the signalby strengthening the signal of interest and restrain the signal of worthless and noises.Based on the adventures above, this article mainly focuses on the adaptive arraybeam-forming in the array signal processing system by introducing new model andalgorithm in connection with the currently existing method during the research, whichaims at the development of robustness and the range of adaption.The main contents ofthe dissertation are described as fellow:1. For the instability problem of calculating the modified signal subspace in thebroad based feature subspace beam-former, this paper puts forward restraining theconstrained subspaces by making use of the dimensions of the block matrix. Thesimulation has showed that the proposed algorithm is better than the LCMV-algorithmwith stable array and there is no need of computing the modified signal subspace. Inthe case when the location of the interruption is unknown, this paper presents usingthe combination of the constraint space to form the constraint space. Both theoreticalanalysis and computer simulation have verified the effectiveness and superiority ofthe algorithm.2. For the large complexity problem of computing the inversion of the samplecovariance matrix, the paper presents to use the characteristics of the generatingcovariance matrix of the Multistage Wiener filter that is the tri-diagonal matrix andmake LU decomposition of the covariance matrix to form the KRYLOV subspaces,which aims at directly get the optimal weights of filter and avoid the iterativeprocess, thus the complexity of the algorithm is greatly reduced.3. For the actual typical adaptive beam-forming DOA mismatch problem, thepaper raises the probability of constraint beam-forming algorithm based on mutativescale chaos optimization in connection with the worst performance of the best soundbeam forming algorithm. By introducing the probability constraints confidence interval and the signal vector error distribution law in the beam-forming, theconstrained optimization problem is eventually turned into a nonlinear optimizationproblem. Then use the ergodicity of the chaotic variables to search global optimumweights.4. Validation and compare of the algorithm is via the MATLAB, by combining theSNR, INR and a random number function to produce the signal model, where K is thesampling number of snapshots. Then set Monte Carlos number to repeat theexperiments in this article, which selecting the number of repetitions of100and500,and making simulation and comparison of the algorithm.