An efficient routing algorithm based on mutual information in unstructured P2P networks

Resource searching is the first and foremost challenge in unstructured P2P networks. Although having received a lot of attentions during past few years, many proposed search methods still can't avoid some serious drawbacks such as high cost and low effciency in dynamic P2P environment. To resolve these weaknesses, this paper proposes an e?cient Query Routing Tree (QRT) algorithm based on maximum mutual information to improve the performance of resource searching, which has tightly associated the resource contents of peers with the logical links of P2P network. The primary benefits are that our proposed QRT algorithm can make the query messages forwarded more effectively in similar peers and hit more target resources faster. The experimental results show that our proposed QRT algorithm can decrease the search cost more effectively and maintain higher targets hit rate compared to Flooding, k-RW and APS algorithms, and the QRT algorithm is also proved to conduct high searching performance with better scalability in unstructured P2P networks. © 2013 Binary Information Press.