摘要
研究提高网络的利用率,在全光网络中放置波长转换器是打破波长一致性约束,为了降低网络阻塞率,提高网络通信能力的有效途径。但限于波长转换器的高昂成本,不可能为网络中的每个节点都配置波长转换器,所以波长转换器应以最优方案放置在网络中的个别关键节点上。提出了一种基于蚁群算法的波长转换器配置方法,通过蚁群算法寻找给定网络中任意源、目的节点之间的最优路径,并利用蚂蚁对最优路径的记录,统计路径在节点处发生波长转换的次数,将具有较高波长转换次数的节点作为网络中波长转换器放置节点。最后通过对一个5节点的网络进行算法演示和仿真分析,结果表明该算法能求得波长转换器的合理配置,得到较好的通信效果。
Wavelength conversion can eliminate the wavelength-continuity constraint and reduce the network blocking probability.Wavelength conversion is the key factor in minimizing the blocking probability and improving network performance in wavelength-routed all-optical networks.This paper proposes an algorithm for wavelength converter allocation in wavelength-routed all-optical networks based on ant colony system,which aims at minimizing the blocking probability and improving network performance.In a given network,in case of all nodes with wavelength converting capability,an optimized path between any source node and destination node would be found by ant algorithms.According to the nodes on the optimized path,the nodes with high wavelength converting rate maintain wavelength converting capability,whiles other nodes will be reset.Simulation results of 5-node network show that this algorithm is reasonable and effective for allocating wavelength converter.
出处
《计算机仿真》
CSCD
北大核心
2010年第1期202-205,270,共5页
Computer Simulation
基金
国家自然科学基金项目(10831006
10671204)
中国科学院知识创新工程重要方向项目(kjcx-yw-s7)资助
关键词
蚁群算法
全光网络
波长转换器
配置
算法
Ant colony system
All-optical networks
Wavelength converter
Placement
Algorithms