Date of Award

12-2013

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Computer Engineering

Advisor

Shen, Haiying

Committee Member

Gemmill , Jill

Committee Member

Wang , Kuang-Ching

Abstract

In peer-to-peer (P2P) file sharing systems, many autonomous peers without preexisting trust relationships share files with each other. Due to their open environment and distributed structure, these systems are vulnerable to the significant impact from selfish and misbehaving nodes. Free-riding, whitewash, collusion and Sybil attacks are common and serious threats, which severely harm non-malicious users and degrade the system performance. Many trust systems were proposed for P2P file sharing systems to encourage cooperative behaviors and punish non-cooperative behaviors. However, querying reputation values usually generates latency and overhead for every user. To address this problem, a social network based trust system (i.e., SocialTrust) was proposed that enables nodes to first request files from friends without reputation value querying since social friends are trustable, and then use trust systems upon friend querying failure when a node's friends do not have its queried file. However, trust systems and SocialTrust cannot effectively deal with free-riding, whitewash, collusion and Sybil attacks. To handle these problems, in this thesis, we introduce a novel trust system, called SocialLink, for P2P file sharing systems. By enabling nodes to maintain personal social network with trustworthy friends, SocialLink encourages nodes to directly share files between friends without querying reputations and hence reduces reputation querying cost. To guarantee the quality of service (QoS) of file provisions from non-friends, SocialLink establishes directionally weighted links from the server to the client with successful file transaction history to constitute a "weighted transaction network", in which the link weight is the size of the transferred file. In this way, SocialLink prevents potential fraudulent transactions (i.e., low-QoS file provision) and encourages nodes to contribute files to non-friends. By constraining the connections between malicious nodes and non-malicious nodes in the weighted transaction network, SocialLink mitigates the adverse effect from whitewash, collusion and Sybil attacks. By simulating experiments, we demonstrate that SocialLink efficiently saves querying cost, reduces free-riding, and prevents damage from whitewash, collusion and Sybil attacks.

Share

COinS