EERP: An enhanced EigenTrust algorithm for reputation management in peer-to-peer networks

Abstract

Open peer-to-peer (P2P) networks have many applications in large-scale distributed systems, internet computing, file sharing and social networks. Peers in P2P networks share data or computing resources while being strangers to each other. In many cases, peers are anonymous volunteers in a dynamic environment where peers can join and leave the network at any time. The open nature of P2P networks is an ideal environment for malicious peers to spread their inauthentic files. Furthermore, some malicious peers form groups to mislead the rest of the peers by giving malicious friends within their groups faked high trust values to increase their reputations. These challenges increase the need to manage the reputation schemes in P2P networks. This paper proposes the EERP algorithm, which focuses on malicious peers and how they can be identified and isolated from the network. This is done by saving the reputation information of each peer as a chain that contains, in addition to the peer’s global reputation value, the rating given to this particular peer after each transaction they are involved in and the identity of the peer who gave the rating. Analysing these logs of peers’ histories helps to track and identify malicious peers. EERP has the potential to effectively reduce the number of inauthentic files spread in a P2P network as compared with other trust management systems.