Differential evolution algorithm for QoS multicast routing

摘要

Computing the multi-constrained least-cost multicast routing tree is an NP-complete problem. In order to solve the quality of service (QoS) constrained routing effectively and efficiently, the scheme of routing based on differential evolution (DE) is proposed. The individuals of the multicast tree which adopt path coding according to the characters of real coding of DE are optimized by mutation and competitive survival strategy. In this paper, the effect of main parameters of DE on the performance of multicast routing algorithm is analyzed. We evaluate the performance and efficiency of the DE algorithm in comparison with the genetic algorithm(GA) by the result of simulation. The DE algorithm is proved to be more probable to find a global optimal solution and to achieve faster convergence.