融合小生境机制的QoS多播路由遗传模拟退火算法

Hybrid genetic simulated annealing algorithm based on niching for QoS multicast routing

针对通信网络中多重QoS约束条件下的多播路由计算,提出了一个基于模拟退火技术的改进遗传算法HGA-QoSR。该算法把模拟退火技术的局部寻优能力与遗传算法的全局寻优能力有机结合,并利用隔离小生境机制控制种群的独立进化,使演化过程中的种群保持生态多样性,以提高算法运行效率和解的质量。理论分析和仿真实验表明,与传统遗传算法相比较,该算法性能有显著改进。

Aiming at the problem of multicast routing with multiple QoS constraint, a new genetic algorithm（GA） was brought up based on the simulated annealing（SA） mechanism, The algorithm combined the capability of local optimization of SA with global optimization of GA, and used of independent evolutionary mechanism of isolation niche. It can effectively enhance the operating efficiency and improve the solution quality. The simulation experiment indicated that, compares with the traditional genetic algorithm, the performance of this algorithm has remarkable improvement.