A Survey of Sybil Attacks in Networks

Most peer-to-peer systems are vulnerable to Sybil attacks. The Sybil attack is an attack wherein a reputation system is subverted by a considerable number of forging identities in peer-to-peer networks. By illegitimately infusing false or biased information via the pseudonymous identities, an adversary can mislead a system into making decisions benefiting herself. For example, in a distributed voting system, an adversary can easily change the overall popularity of an option by providing plenty of false praise, or bad-mouthing the option through these fake identities. In this paper, we summarize the existing Sybil defense techniques, and further provide some new research areas. Unlike traditional surveys about Sybil defense, we first categorize the Sybil defense methods, mainly according to their designed time, and then classify the methods by their approaches. We believe that by understanding the evolution of the solutions, readers could essentially have more insights on the problem. In a nutshell, the research on the Sybil defense technique has experienced four phases: (1) traditional security key-based approaches, (2) specific peer-topeer system feature-based solutions, (3) social network-based methods, and (4) social community-based techniques. Besides all of these anti-sybil methods, readers will also find some Sybil attack-related topics, such as Sockpuppets in online discussion forums. By the end of the paper, we will provide some predictions about directions for future research.