【摘要】 节点选择机制是P2P网络的研究热点与核心技术,近年来得到广泛的关注.该文将一种改进的粒子群优化算法-具有双重结构编码的二进制粒子群优化算法(DS-BPSO)应用于P2P文件共享系统的邻居节点选择过程.在动态性方面,粒子群优化算法(PSO)与P2P网络具有先天的相似性.每个粒子(节点)通过有向图来编码与其它粒子的连接矩阵,更符合ADSL连接常规.实验结果表明:对于大规模的邻接选择问题,DS-BPSO算法通常会比二进制粒子群优化算法(BPSO)能够查找到更好的邻居节点.更多还原

【Abstract】 Node selection mechanism is the research focus and the core technology in P2P network,recent years it has been widely concerned.In this article,a modified Particle Swarm Optimization Algorithm——double structure encoded Binary Particle Swarm Optimization Algorithm(referred as DS-BPSO) will be used in the process of adjoining selection for P2P file sharing system.In term of dynamics,Particle Swarm Optimization Algorithm(referred as PSO) has inherent similarities with the P2P network.Each particle(i.e.node) encodes the connection matrix with other particles through the directed graph,which will more match the connection routine of ADSL in real life.Experiment results indicate that,DS-BPSO algorithm usually could obtain better neighbor nodes than the Binary Particle Swarm Optimization Algorithm(referred to as BPSO),especially for large scale problems.更多还原