一种基于计算智能的组播路由算法

Multicast Routing Algorithm based on Computational Intelligence

在通信网络中,多约束组播通信是提高网络运行效率和服务质量的重要途径。一些启发式的算法已经被用来解决多约束条件下的组播路由问题,如模拟退火算法,遗传算法,蚁群算法和粒子群优化算法等。然而,这些算法在求解多约束组播路由问题时存在收敛速度低和计算复杂度高的问题。萤火虫群优化(GSO)算法是一种近期在计算智能领域出现的卓越算法,它可以在一定程度上解决多约束组播树生成过程中收敛速度低和计算复杂度高的问题。提出了一种基于GSO的多约束组播树生成算法(GSO-MCM)。该算法可有效生成满足多约束要求的组播路由树。仿真结果表明提出的GSO-MCM算法在求解和收敛速度,以及网络规模适应性方面均有良好的性能。

In communication networks, the multi-constraint muhicast communication is an important way to improve the e^ciency of network operation and QoS （Quality of Service）. Some heuristic algorithms are applied to solving muhicast routing problem under multiple constraints, such as simulated annealing, ge- netic algorithm, ant colony algorithm and particle swarm optimization algorithm. However, problems of low convergence rate and high computational complexity still exist in these algorithms when solving muhi-con- straint multicast routing problems. GSO （Glowworm Swarm Optimization） algorithm is a promising algo- rithm recently emerging in the area of computational intelligence, and it can overcome the above deficien- cies to some extent. Meanwhile, GSO-MCM algorithm based on GSO is proposed to efficiently generate multieast routing tree and meet multi-constraint requirements. Simulation result shows that GSO-MCM al- gorithm enjoys good performance in solution, rate of convergence and adaptability of network size.