计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2014年
12期
266-270
,共5页
遗传算法%模拟退火算法%波达方向(DOA)
遺傳算法%模擬退火算法%波達方嚮(DOA)
유전산법%모의퇴화산법%파체방향(DOA)
genetic algorithm%simulated annealing genetic algorithm%Direction Of Arrival(DOA)
将模拟退火思想融入到遗传算法中，形成了另一种优化算法，即模拟退火遗传算法，将其应用于加权子空间（WSF）算法的目标方位（DOA）估计技术中，以求降低WSF算法的运算复杂度，提高DOA估计精度，同时又解决了基本遗传算法在DOA估计中易陷入局部最优、后期搜索迟钝等问题。计算机仿真结果表明：采用模拟退火遗传算法的DOA估计技术在低信噪比条件下比采用基本遗传算法、高斯-牛顿算法有更高的分辨概率，更小的均方误差。
將模擬退火思想融入到遺傳算法中，形成瞭另一種優化算法，即模擬退火遺傳算法，將其應用于加權子空間（WSF）算法的目標方位（DOA）估計技術中，以求降低WSF算法的運算複雜度，提高DOA估計精度，同時又解決瞭基本遺傳算法在DOA估計中易陷入跼部最優、後期搜索遲鈍等問題。計算機倣真結果錶明：採用模擬退火遺傳算法的DOA估計技術在低信譟比條件下比採用基本遺傳算法、高斯-牛頓算法有更高的分辨概率，更小的均方誤差。
장모의퇴화사상융입도유전산법중，형성료령일충우화산법，즉모의퇴화유전산법，장기응용우가권자공간（WSF）산법적목표방위（DOA）고계기술중，이구강저WSF산법적운산복잡도，제고DOA고계정도，동시우해결료기본유전산법재DOA고계중역함입국부최우、후기수색지둔등문제。계산궤방진결과표명：채용모의퇴화유전산법적DOA고계기술재저신조비조건하비채용기본유전산법、고사-우돈산법유경고적분변개솔，경소적균방오차。
The simulated annealing genetic algorithm is a new global optimization algorithm, and it is formed by integrating the simulated annealing into the genetic algorithm. Then the simulated annealing genetic algorithm is applied to the WSF algorithm of DOA estimation technique, in order to reduce the complexity of WSF algorithm and improve the DOA estima-tion precision. At the same time, the new algorithm can solve the low efficiency and easily falling into local optimum prob-lems of the basic genetic algorithm in DOA estimation. Computer simulation results show that, compared with the basic genetic algorithm, gauss-newton method, the DOA estimation technique based on simulated annealing genetic algorithm has higher resolution probability and smaller mean square error.
