Supported by the National Natural Science Foundation of China under Grant Nos.60073053, 60133010 (国家自然科学基金); the National High-Tech Research and Development Plan of China under Grant No.2002AA135080 (国家高技术研究发展计划(863))
近年来,时延受限的代价最小多播树问题备受关注.到目前为止,BSMA(bounded shortest multicast algorithm)算法被认为是最好的受限多播路由算法;然而,过长的计算时间限制了其应用.作为一种全局优化算法,遗传算法(genetic algorithm,简称GA)被越来越多地应用于多播路由问题.与传统的算法相比,遗传算法的全局搜索能力更强,但其易"早熟"的特点使它并不总是能得到最优多播树.提出的基于克隆策略的多播路由算法,有效地解决了"遗传"多播路由算法中的"早熟"问题,并通过引入一个可调因子缩小了搜索空间,加快了算法的收敛速度.算法实现简单、控制灵活.仿真结果表明,该算法的性能优于BSMA算法和传统的遗传算法.
The problem of computing delay-constrained minimum-cost multicast trees is of great interest in the last few years. So far, the Bounded Shortest Multicast Algorithm (BSMA) has been thought to be the best constrained multicast algorithm. However, the large computation time restricts its application. As a global optimizing algorithm, Genetic algorithm (GA) is applied to solve the problem of multicast more and more. GA has more powerful searching ability than traditional algorithm, however, the property of "prematurity" makes it difficult to get a good multicast tree. A Clonal Strategies (CS) based multicast algorithm is presented in this paper, which saliently solves the "prematurity" problem in Genetic based multicast algorithm. Furthermore, the algorithm is accelerated by using an adjustable parameter to reduce the search space. The algorithm has the property of simple realization and flexible control. The simulated results show that CS has better performance than BSMA and GA.