蚁群算法在QoS单播路由中的应用研究  

Application Research of Ant Colony Algorithm for QoS Unicast routing

作　　者：岐世峰[1] 李艳华[1] 梅大成[2] 

机构地区：[1]攀枝花学院计算机学院,四川攀枝花617000 [2]西南石油大学计算机学院,成都610500

出　　处：《电子科技大学学报》2010年第2期271-274,共4页Journal of University of Electronic Science and Technology of China

基　　金：四川省科技厅应用技术研究项目基金(2008JY0114)

摘　　要：QoS路由问题被证明是一个NP-C问题,而传统的路由算法很难有效地解决NP-C问题。该文提出了一种基于蚁群算法、用于解决带宽和时延约束问题的QoS单播路由算法,利用蚁群算法中蚂蚁通过信息素寻找最优路径的机制,并以网络吞吐量和数据报的平均时延等性能为最优的准则,来定义蚂蚁的转移概率、路由表和信息素更新方式,实现基于蚁群算法的路由选择算法.这种算法具有较强全局最优解搜索能力,较强的灵活性,以及潜在的并行性。QoS routing problem is proved to be a NP-C problem, it is very difficult to solve the NP-C problem effectively by conventional routing algorithms. This paper puts forward an QoS unicast routing algorithm which is based on the principle of ant colony algorithm and used to solve the problem of bandwidth and delay constrain. The rule of ant colony algorithm that the ants find the shortest path through the laying down of pheromone and the rule with emphasis on the maximal network throughput and the lowest average cell delay are also used to define the transition probability of ants, routing table, and the way pheromone is updated. A detailed routing scheme algorithm based on the ant colony algorithm is implemented. Results prove that the algorithm has strong ability of searching an optimum solution and show its better flexibility and potential parallelism.

关 键 词：蚁群算法 信息素 QOS路由 路由算法 转移概率 

分 类 号：TP393.0[自动化与计算机技术—计算机应用技术]