DOA Estimation Based On The Improved Population Genetic Algorithms

MUSIC algorithms as a classic super-resolution algorithm for spatial spectrum estimation has important application in the multi-dimensional parameter estimation of array signal. The MUSIC algorithm can be divided into eigenvalue decomposition and spectrum peak search, because the calculation of searching spectral peak process is heavy,it can't meet the requirement of high real-time. GA (Genetic Algorithm) is a bionic algorithm which has the advantages of parallelism,global optimization and simplicity etc. But it also has some problems. In this paper, we improved the genetic algorithms to the premature convergence phenomena and the problem of global search ability.In this paper, On the basis of the genetic algorithm, An improved population bee evolutionary genetic algorithms (IPGA) is proposed. In the algorithms, the best chromosome among the current population is crossovered with drones selected according to a certain crossover probability,which enhances the exploitation of searching global optimum. A local search is adopted to randomly produce more parents around the secondary best chromosome, so as to escape the premature and increase the exploration ability at the same time, the algorithm uses the strategy by stages to adjust the dynamic random population scale. The experimental results prove that the proposed algorithm is actually a great improvement of genetic algorithms in solution precision.Finally, the improved genetic algorithms is applied to the MUSIC algorithms spectral peak searching process. The performance of the improved genetic algorithms for the music spectrum peaks in the case of single-source and multi-source search are simulated. The simulated result shows that it has high efficiency and strong capability of global search. Compared genetic algorithms and the traversal algorithms, genetic algorithms can largely reduce the computation of the MUSIC spectrum peak search.This study not only improve the real-time performance of the MUSIC algorithms,also has important reference value for other complex function optimization.