Hybridized Optimization Genetic Algorithm for QOS-Based Multicast Routing Problem

This paper analyzes the mathematic model of QOS(quality-of-service) based multicast routing with nodes delay. Based on the original, an optimization model called GP-GA is proposed by hybridizing Gene-Pool (GP) with traditional Genetic Algorithm (GA). The model uses the gene-pool to save the better individual during the process of each generation. In the mean time, the crossover and mutation operator are improved to accelerate the convergence. As the problem may be easily trapped by local optimization, the evolution strategy based on 'reserved and non-reserved' is also constructed to enhance the ability of finding optimal solution and decrease the probability of 'premature' phenomena commendably. The emulation experiments demonstrate that the probability of GP-GA is higher than the general GA in converging optimal solutions, and the algorithm is also effective for the adjustment to the dynamic multicast routing.