QoS多播路由算法研究  

Method for multicast routing based on genetic algorithm

作　　者：杨德红[1] 瞿中[2] 何江平[3] 

机构地区：[1]重庆工学院计算机科学与工程学院,重庆400050 [2]重庆邮电学院计算机科学与技术学院 [3]重庆工学院数理学院,重庆400050

出　　处：《计算机工程与设计》2005年第10期2730-2733,共4页Computer Engineering and Design

摘　　要：随着当前Internet的发展和各种多媒体应用的出现,多播技术得到大量应用。多播路由算法主要用来建立一棵性能良好的多播树,并使它能够满足各种业务的服务质量需求。将多种群并行技术和退火技术相结合,克服了基于标准遗传算法的多播路由算法过早收敛和后期搜索速度较慢的缺陷,且使用树状编码方法,提出求解带宽、时延、时延抖动和分组丢失率约束的代价最小多播树的多种群并行退火遗传多播路由算法。对QoS多播路由选择问题进行了描述,给出多种群并行退火多播路由遗传算法和一种有效去除冗余信息的遗传算法编码设计技术,通过仿真实验证明了算法的正确性,分析了算法的时间性能,表明该算法快速有效。With the development of Intemet and the advent of various multimedia applications, multicasting technology is widely applied. Multicast routing constructs paths along with data packets from a source were distributed to reach many, but not all, destinations in a communication network. In order to overcome the pre-maturity and low speed of search in the late phase ofmulticast routing algorithm based on standard genetic algorithm, the multi-population parallel annealing genetic multicast routing algorithm was given to solve the bandwidth, delay, delay jitter and packet loss constrained least-cost multicast routing problem, which combined the multi-population parallel technology and annealing technology and adopted tree-like coding approach. The problem ofQoS multicast routing was described, a multi-population armealing multicast routing genetic algorithm （MPAMRGA） and a encoding method based on genetic algorithm eliminating redundancy were provided. The simulation results suggest that the algorithm is feasible and effective.

关 键 词：QOS 多播 遗传算法 编码 

分 类 号：TP393.03[自动化与计算机技术—计算机应用技术]