Abstract:

In peer-to-peer network, there exist lots of free riders, whose passive behaviors decrease the efficiency and reliability of search. A new search algorithm named credit model based restricted random walk (CMRRW), which is based on the credit model, is proposed to solve this problem. The credit model is applied to evaluate the reliability of the peers. The query reliability rate is gained to reflect the peer’s history query success rate and the routing efficiency of its neighbors. In our solutions, when the peer selects neighbors to forward the query message, the neighbors with higher credit and history hit rate will be selected in prior order. Simulation results indicate that CMRRW shows good performance.

Key words: peer-to-peer network, random walk, incentive mechanism, search