Parameter adjustment based on improved genetic algorithm for cognitive radio networks  

作　　者：ZHAO Jun-hui LI Fei ZHANG Xue-xue 

机构地区：[1]State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, China [2]Key Laboratory of Wireless Sensor Network & Communication, Shanghai Institute of Mierosystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2012年第3期22-26,共5页中国邮电高校学报（英文版）

基　　金：supported by the National Natural Science Foundation of China (61172073);National Key Special Program(2012ZX03003005);the State Key Laboratory of Rail Traffic Control and Safety (RCS2011ZT003);Beijing Jiaotong University and the Fundamental Research Funds for the Central Universities

摘　　要：Multi-objective parameter adjustment plays an important role in improving the performance of the cognitive radio （CR） system. Current research focus on the genetic algorithm （GA） to achieve parameter optimization in CR, while general GA always fall into premature convergence. Thereafter, this paper proposed a linear scale transformation to the fitness of individual chromosome, which can reduce the impact of extraordinary individuals exiting in the early evolution iterations, and ensure competition between individuals in the latter evolution iterations. This paper also introduces an adaptive crossover and mutation probability algorithm into parameter adjustment, which can ensure the diversity and convergence of the population. Two applications are applied in the parameter adjustment of CR, one application prefers the bit error rate and another prefers the bandwidth. Simulation results show that the improved parameter adjustment algorithm can converge to the global optimal solution fast without falling into premature convergence.Multi-objective parameter adjustment plays an important role in improving the performance of the cognitive radio （CR） system. Current research focus on the genetic algorithm （GA） to achieve parameter optimization in CR, while general GA always fall into premature convergence. Thereafter, this paper proposed a linear scale transformation to the fitness of individual chromosome, which can reduce the impact of extraordinary individuals exiting in the early evolution iterations, and ensure competition between individuals in the latter evolution iterations. This paper also introduces an adaptive crossover and mutation probability algorithm into parameter adjustment, which can ensure the diversity and convergence of the population. Two applications are applied in the parameter adjustment of CR, one application prefers the bit error rate and another prefers the bandwidth. Simulation results show that the improved parameter adjustment algorithm can converge to the global optimal solution fast without falling into premature convergence.

关 键 词：cognitive radio genetic algorithm global optimal solution linear scale transformation adaptive crossover and mutation probability 

分 类 号：O224[理学—运筹学与控制论] TN929.5[理学—数学]