Abstract: This paper studies search mechanism in unstructured Peer-to-Peer(P2P) network and introduces an improved algorithm based on Bloom Filter(BF). The algorithm uses BF technology to generate routing item and exchanges local routing table in limited range, which makes nodes know others’ shared information in this range. So it can realize purposive search and avoid traditional blind search. Simulation results show that the amount of message created by queries of this algorithm is one order of magnitude lower than traditional search and this algorithm can gain better recall rate.

Key words: Peer-to-Peer(P2P) network, Bloom Filter(BF), routing, search

摘要： 研究非结构化P2P网络的搜索机制，提出基于布莱姆过滤器(BF)路由表的改进算法。该算法利用BF技术生成路由条目并在一定范围内相互交换本地路由表，使节点能够了解一定范围内的节点共享信息，实现有针对性的搜索，避免传统的盲目性搜索。仿真结果表明，该算法查询搜索时产生的消息数量比传统算法减少一个数量级，并能够获得较好的查全率。

关键词: 对等网络, 布莱姆过滤器, 路由, 搜索

CLC Number: 

• TP393