P2P网络中基于蚁群算法的资源搜索研究

实际的P2P网络中包含特定资源的节点只占一部分,而多数搜索方法大多都会遍历整个网络节点,节点的动态性又极易造成搜索路径失效.因此提出了一种基于蚁群算法的搜索机制,该机制以节点上资源访问成功率来定义节点信息素,以避免对消极节点的访问,以节点间的通信次数定义相邻节点的路径信息素,以避免对失效路径的使用,同时结合路径长度对被选路径进行优化选择.给出了该机制的详细搜索算法,并进行了仿真实验.实验结果表明,与Flooding和Chord算法相比,该算法在提高搜索成功率的同时也降低了查询次数.

Research on search mechanism based on ant colony algorithm in P2P networks

In the actual P2P networks,only few peers contain search resources,and most search methods will traverse the entire network peers.In addition,the dynamic nature of network peers is easy to make the search path useless.According to these questions,a search mechanism based on ant colony algorithm is introduced,in this search mechanism,peer-pheromone is defined based on peer’s success percent of resources contributions to avoid visits of passive node,besides,path-pheromone is also defined based on number of communications to avoid useless path.In the while,this mechanism makes finally choose among the already acquired paths.Detailed steps of algorithm and a simulation experiment are offered.Experimental results show that this algorithm can not only reduce the number of routing query,but also greatly improve the search hit ratio compared with Flooding and Chord.