PeerMonitor: A P2P-based architecture for video monitoring

With rapidly growing demand for real-time video monitoring, providing a robust large-scale video monitoring service is a both challenging and crucial task. This paper proposes a P2P-based video monitoring architecture called PeerMonitor which can implement large-scale network video monitoring application using a relatively small number of server. By fully taking different use interests in different monitoring scene into account, PeerMonitor provides a more efficient scene switch operation. The proposed architecture also decreases video delivering delay by improving Pure-Pull strategy and data-flow-based push-pull strategy between nodes in DONet. The simulation result demonstrates the efficiency of the proposal.