Evaluation of Index Poisoning Method in Large Scale Winny Network

In recent years, P2P file-sharing networks are used all over the world. This has led to social problems such as the illegal distribution of copyrighted material and the leakage of personal information through computer viruses because P2P does not have a control function for file distribution. To address these issues, a control method called index poisoning has been studied. However, problems such as pollution in the network index and a need to increment control traffic have been reported when index poisoning is applied to a P2P file-sharing network. We propose a method that implements dynamic clustering to limit the range of index poisoning as a solution for these problems that also maintains the effectiveness of the control function. In this study, we implement index poisoning on the large-scale Winny network and evaluate the system performance.