基于信息熵多种群遗传算法的组播路由

Multicast Routing Based on Information Entropy Multi-population Genetic Algorithm

针对网络信息不确定性和链路负载不均匀所造成的网络拥塞,提出基于信息熵的组播路由算法。该遗传算法从最小代价树开始,在多种群中不断选择信息熵较大的种群,以求得满足延时要求且路径负载较小的组播树。结果表明,该算法性能快速、有效地构造最小时延组播树,且保证网络负载均衡分布。

Taking account of the uncertain information and the unbalance of link results in the congestion of network, a Genetic Algorithm（GA） with the information entropy is presented to solve the problem in the multicast routing. The algorithm begins from a least-delay tree, searches the max information entropy multicast tree in the multi-population, and gets final multicast tree satisfying delay constraint and the least balance. The results show that the proposed algorithm performs better in terms of delaying a running time against existing heuristics algorithm, and constructs optimal delay-constrained and balance multicast tree efficiently.