一种针对频率分配问题的改进ANTS算法  

An Improved ANTS Algorithm for Frequency Assignment Problem

作　　者：徐奇[1,2] 熊晖[1] 李钊[2] 陈大勇[2] 

机构地区：[1]国防科技大学,湖南长沙410073 [2]第二炮兵驻石家庄地区军事代表室,河北石家庄050081

出　　处：《无线电工程》2010年第1期58-61,共4页Radio Engineering

摘　　要：频率分配问题是近年来通信领域研究的热点。针对FAP问题提出了一种结合模拟退火算法的改进ANTS算法。运用模拟退火算法产生次优解,利用次优解分配初始信息素,并利用ANTS算法来寻求最佳方案。在ANTS算法的每个蚂蚁寻找局部最优过程中,为了加快运算速度,对局部寻优过程进行了改进。实验结果表明,在解质量相当的情况下,该算法能够大大地加快收敛速度,特别是针对一些较复杂的分配情况,效果明显。Frequency assignment problem （FAP） catches many scholar＇s eyes. A new method combined with simulated annealing algorithm is proposed in the article for the FAP. Firstly a simulated annealing algorithm procedure is used to solve the problem to get a second- best solution. Secondly the solution is used to distribute the original pheromone. And ANTS algorithm is used to search the optimal solution. While each ant is constructing the solution, we improved the local search procedure to accelerate the convergence time. Experimental results show that the algorithm can accelerate the convergence speed and attain the same conclusion, especially for some complex problem.

关 键 词：频率分配 ANTS 模拟退火算法 局部寻优 

分 类 号：TN915.01[电子电信—通信与信息系统]