A Novel Clustering Tree-based Video lookup Strategy for Supporting VCR-like Operations in MANETs
Shijie Jia, Ye Guo, Youzhong Ma, Chunlin Kuang, Guofang Kuang
Mobile Peer-to-Peer (MP2P) technologies can efficiently support large-scale deployment for video streaming services over mobile ad-hoc networks (MANETs) such as video-on-demand (VoD). Because streaming interactivity for VoD service causes frequent video lookup, the video lookup performance of MP2P-based VoD systems is significant for system quality of service (QoS) and user quality of experience (QoE). In this paper, we propose a novel Clustering Tree-based Video Lookup strategy for supporting VCR-like operations in MANETs (CTVL). CTVL builds the binary tree composed of chunks with high popularity by investigation of playback and switchover frequency of users for video chunks. Based on the built binary tree, CTVL designs a construction method of node community to group nodes into different communities in terms of cached chunks and defines the logical links between communities, which reduces maintenance cost of community structure and achieves high system scalability. CTVL also designs a video chunk lookup strategy, which makes use of the logical links between communities to achieve fast video chunk lookup. Extensive tests show how CTVL achieves much better performance results in comparison with other state of the art solutions.
Kata kunci : Keywords: binary tree, video chunk, communityView : 159 | Download : 70