多点并行蚁群搜索在多限制动态组播中的应用研究  

Dynamic Multi-constrained Distributed Multicast Routing Algorithm Based on Multi-point Parallel Ant Search

作　　者：宋学军[1] 刘巍[1] 

机构地区：[1]天津大学电气自动化与能源工程学院,天津300072

出　　处：《电路与系统学报》2004年第1期73-77,118,共6页Journal of Circuits and Systems

基　　金：教育部博士学科点基金资助项目(200005634)

摘　　要：随着Internet的快速发展和商用化的提高,多媒体组播通信越来越引起人们的广泛关注。以往的研究成果大多是集中在固定的多点模型上,而许多服务是无法应用这种应用模型的。本文在充分研究以往组播通信的基础上,针对没有固定组成员的动态组播通信提出了运用蚁群算法进行并行寻路的算法模型。同时,为了加快蚁群算法的收敛速度,对该算法进行了一定的改进。仿真试验表明,该算法是一种有效的适应动态组成员的组播路由算法,具有一定的实用价值。With the rapid progress of the commercial use in internet, multi-media multicast application attracts more research attention. Currently used multicast model are base on multiple fixed points. However, this static model can not adapt many new services, such as VOD or tele-medical applications, because users of these services access the Internet randomly. What we need is a dynamically changing multicast network with dynamic changing constrains (bandwidth requirement). Based on the extensive investigation of existing multicast algorithms, a dynamic multicast model based on ant search routing algorithm is proposed. Further improvement is also made to accelerate the speed of convergence. Results of simulation demonstrate that the proposed model is effective to dynamic group member multicast routing algorithm. Therefore, It is promising to provide practical multicast services.

关 键 词：动态组播 蚁群算法 并行运算 

分 类 号：TN913[电子电信—通信与信息系统]