P2P Streaming Method Based on Playback Deadline Using Linear Programming

Abstract

In this study, we propose a Peer-to-Peer (P2P) Video-on-Demand (VoD) streaming method which adjusts the receiving rate of pieces by considering the playback deadline of the pieces to reduce interrupted playback. P2P VoD streaming methods based on BitTorrent, such as BitTorrent Streaming (BiToS) and BiToS + Immediacy and Scarcity (BIS), have been proposed to support streaming. In the piece selection strategies of these two methods, a peer probabilistically selects a piece considering immediacy and/or scarcity (i.e., rarity) and receives the selected piece to reduce interrupted playback. However, the piece selection mechanisms still cause interrupted playback owing to probabilistic piece selection. In our proposed P2P VoD streaming method, a peer sequentially selects a piece, starting from the first piece and receives the selected piece at a transfer rate in time for predicted playback start time. As a result, the interrupted playback can be reduced further. In evaluation simulations, we show that the proposed streaming method outperforms previous streaming methods in terms of media playback continuity.

Application Layer Multicast Based Services on Hierarchical Peer to Peer Architecture

“…NICE [9] uses hierarchical clustering techniques to build overlay tree, whereby group members arrange themselves into clusters with the neighboring nodes. Recent type of ALM based applications have been emerged the P2P networking such as streaming [14], telemedecine [15] and social networking [16]. The main objective of our proposed model is to build an efficient and optimized multicast tree from any node (root or sender) to all other nodes (receivers) using HPM as underlying architecture.…”