Refining Reputation to Truly Select High-QoS Servers in Peer-to-Peer Networks

摘要: Peer-to-peer networks (P2Ps) use reputation systems to provide incentives for nodes offer high quality of service (QoS) and thwart the intentions dishonest or selfish nodes. Existing have two problems. First, they directly regard node as trust. Rather, represents opinions formed by others about a node's QoS behavior, while trust honesty willingness cooperate. In addition trust, factors such capacity lifetime also influence reputation. Due these factors' heterogeneity variance over time, cannot reflect current QoS. Second, existing guide select server with highest reputation, which may not actually would overload reputed This work aims accurately guidance high-QoS selection. Through experimental study, we find that available capacity, positively affect Based on this observation, first propose manual model an automatic remove additional truly We then selection algorithm separately considers lifetime. Extensive simulation results demonstrate effectiveness models in accurate reflection compared system. Moreover, dramatically increases success rate requests avoids overloading