AntiLiar: Defending against cheating attacks in mesh based streaming

Peer-to-peer (P2P) mesh based streaming systems have gained widespread use for multicasting of audio and video. In approaches based on BitTorrent, members of the swarm share their content availability through gossiping, and redistribute file pieces cached locally. However, such systems are vulnerable to cheating attacks such as fake reporting, selective omission, fake block attack, and neighbor selection attack. These attacks will severely impact quality of service, waste resources, and discourage cooperation among participants. A defense mechanism, called AntiLiar, is proposed for defending against cheating attacks in mesh based streaming. AntiLiar uses a secure progress log that consists of commitments and one-time signatures, and maintains consistency for an expanded neighbor view. Experimental results demonstrate that AntiLiar minimizes required costs, and improves service quality over alternatives in the face of cheating attacks.