摘要
原有启发式P圈构造算法Grow没有考虑到在对基本圈上的一条边扩张时可能会抑制其他边的扩张,从而失去一些先验效率较高的P圈。针对这一缺点对Grow算法进行了改进。改进后的算法根据所有的圈上边扩张后得到的P圈先验效率的高低顺序,优先选择引入的P圈具有最高先验效率的边进行扩张,因此每次扩张都最大可能地提高P圈的先验效率,而且P圈平均跳数基本不变。仿真结果表明,此算法可以在减少备选P圈数量的同时,提高网络的资源利用率。
Since the existing heuristic algorithm named Grow did not consider that expansion on a cycle-link might restrict ex- pansion on another one, therefore some P-cycles with high AE might be lost. For this reason, this paper improved Grow algo- rithm. Among all the on-cycle spans, the improved algorithm first chose the one which could get the P-cycle with the highest AE after expansion, so AE of P-cycle would be increased as high as possible, and average number of hops was more or less the same with Grow＇ s. Finally, the simulation results show that the algorithm can reduce the number of candidate P-cycles, and at the same time achieve better capacity utilization than Grow.
出处
《计算机应用研究》
CSCD
北大核心
2009年第9期3219-3221,共3页
Application Research of Computers
基金
国家自然科学基金面上项目(60778007)
关键词
光网络
保护
生存性
预置圈
optical network
protection
survivability
preconfiguration cycle
