摘要
传统的遗传算法在解决认知无线电频谱分配问题时,没有考虑染色体中来自于不同频谱的基因所表达的遗传特性是不同的,而不加区别地对染色体进行交叉会降低其进化效率。针对此问题,依据遗传特性把染色体分成不同的片段,将染色体交叉设定在每一个片段内,并加入了染色体片段重组过程,用来提高染色体进化的效率。然后从系统公平性的角度设计了自适应的变异概率,让接入率较低的染色体获得更大的变异机会,以提高系统的公平性。最后与遗传算法(genetic algorithm,GA)和量子遗传算法(quantum genetic algorithm,QGA)进行了仿真对比实验,结果表明该算法的收敛速度更快,且同时获得了较高的系统效益以及用户接入率。
Traditional genetic algorithm in solving the problem of cognitive radio spectrum allocation,without considering the genetic characteristics of genes expressed from the chromosome in different spectrum is different,and indiscriminate cross-chromosome evolution will reduce its efficiency. To solve this problem,this paper based on genetic characteristics divided the chromosome into different segments,chromosome crossover would be set within each segment,and joined the chromosome segment restructuring process,to improve the efficiency of chromosomal evolution,and from the perspective of fairness design adaptive mutation probability,so that the lower rate of chromosomal gain access to greater mutating opportunities to improve the fairness of the system. Finally with genetic algorithm and quantum genetic algorithm for the simulation experiments comparing,the results show that this algorithm converges faster,and also wins the higher system benefit and user access ratio.
出处
《计算机应用研究》
CSCD
北大核心
2016年第4期1219-1223,共5页
Application Research of Computers
基金
国家自然科学基金资助项目(61379057
61309001
61379110
61103202)
国家"973"计划资助项目(2014CB046305)
国家教育部博士点基金新教师类资助项目(20110162120046)
中南大学中央高校基本科研业务费专项资金资助项目(2015zzts232)
关键词
认知无线网络
频谱分配
遗传算法
系统效益
用户接入率
cognitive radio network
spectrum allocation
genetic algorithm
system benefit
user access ratio
