PIFF: An Intelligent File Filtering Mechanism for Peer-to-Peer Network

The distribution of copyright content over peer-to-peer (P2P) networks facilitates misuse of protected digital property and severely violates the rights of creators and owners. Besides, the anonymity and other features of latest P2P applications make law enforcement almost impractical. Unfortunately, current technology efforts attempting to block unauthorized files depend on manually predefined keywords which can be easily modified by the application, and they fail to distinguish between authorized and unauthorized files. In this context, we present PIFF, an intelligent file filtering mechanism for peer-to-peer network, which effectively blocks illegal file distribution based on file signatures automatically generated and deployed by the system. With a hierarchy infrastructure and decision algorithm, our solution discovers and identifies the file signatures much faster than manual analysis, and spreads it among system nodes to perform filtering immediately. Experiment results suggest that compared with unprotected environment, PIFF can reduce over 85% peers for unauthorized files without any affect on the distribution of authorized files