基于遗传算法的多目标QoS多播路由算法  

A new multi-objective algorithm of QoS-based multicast routing based on genetic algorithm

作　　者：徐磊[1] 武坤[1] 

机构地区：[1]中南大学数学科学与计算技术学院,湖南长沙410083

出　　处：《电脑与信息技术》2006年第6期12-15,共4页Computer and Information Technology

摘　　要：针对通信网中多目标QoS多播路由模型,文章提出了一种改进的遗传算法。算法的编码方式采用节点序编码,编码和解码过程都比较简单直观。采用保留最佳个体与适应度比例的选择方法,加快了遗传算法的收敛,并在此基础上采用理想点法构建评价函数来求解模型的Pareto解,采用双层的交叉与变异操作,增加群体的多样性。仿真试验验证,该算法可靠性高,适用于大规模网络。This paper brings forward a new algorithm which improved for the QoS-based Multicast Routing Model. The algorithm adopts the integral serial in coding and decoding of the path. It uses best individual reserving and fitness proportional methods to speed convergence and builds the fitness function on the ideal point to get the Pareto solution, It adopts dual layer crossover and mutation operator to make population varied. The simulation results show the algorithm is high reliability. It can be used to find the solution for a large scale network.

关 键 词：遗传算法 QoS 多播路由 多目标 PARETO最优解 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]