Informationally Decentralized Video Streaming Over Multihop Wireless Networks
IEEE transactions on multimedia
Multi-user video streaming over multi-hop wireless networks has recently emerged as not only an important application, but also as an active and challenging research area, as the stringent delay requirements imposed by real-time multimedia transmission need to be satisfied despite the limited-bandwidth, time-varying and error-prone network infrastructure. Various packet scheduling, dynamic routing, error-protection and channel adaptation strategies have been proposed at different layers of the
... rent layers of the protocol stack to address this problem. However, these cross-layer transmission strategies can be efficiently optimized only if they use accurate information about the network conditions and hence, are able to timely adapt to network changes. Due to the informationally decentralized nature of the multi-hop wireless network, performing centralized optimization for delay-sensitive video streaming application based on global information about the network status is not practical. Distributed solutions that adapt the transmission strategies based on timely information feedback need to be considered. To acquire this information feedback for cross-layer adaptation, we deploy an overlay infrastructure, which is able to relay the necessary information about the network status and incurred delays across different network "horizons" (i.e. across a different number of hops in a predetermined period of time). In this paper, we propose a distributed streaming approach that is optimized based on the local information feedback acquired from the various network horizons. We investigate the distributed cross-layer adaptation at each wireless node by considering the advantages resulting from an accurate and frequent network information feedback from larger horizons as well as the drawbacks resulting from an increased transmission overhead. Based on the information feedback, we can estimate the risk that packets from different priority classes will not arrive at their destination before their decoding deadline expires. Subsequently, the various transmission strategies such as packet scheduling, retransmission limit and dynamic routing policies are adapted to jointly consider the estimated risk as well as the impact in terms of distortion of the different priority classes. In our results, we show the impact of the various information horizons in different network conditions. Moreover, our results show that the proposed dynamic routing policy based on timely information feedback outperforms existing state-of-the-art on-demand routing solutions by more than 2dB in terms of the received video quality. This analysis and experiments provide important insights for future protocols and system design targeted at enhanced multi-user video streaming support across wireless multi-hop networks. Index Terms: multi-user video streaming, video packet scheduling, decentralized information feedback, multi-hop wireless networks, dynamic routing.