TRADER: A reliable transmission scheme to video conferencing applications over the internet

• Published in: Journal of network and computer applications Vol. 44; pp. 161 - 171
• Main Authors: Wu, Jiyan, Shang, Yanlei, Yuen, Chau, Cheng, Bo, Chen, Junliang
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.09.2014; Elsevier
• Subjects: Access methods and protocols, osi model; Applied sciences; Broadcasting. Videocommunications. Audiovisual; Burst loss; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Computer systems performance. Reliability; Exact sciences and technology; Forward error correction; Software; Telecommunications; Telecommunications and information theory; Teleprocessing networks. Isdn; Transmission delay differences; Video conferencing; Videocommunications (networks and services); Transmission delay differences; Forward error correction; Video conferencing; Burst loss; Performance evaluation; Continuous time; Distributed computing; Modeling; Videoconference; Optimization; Research and development; Delay; Markov chain; Image coding; Multimedia; High performance; Packet switching; Redundancy; Teleconference; Real time; Transmission time; Image quality; Experimental result; Traffic management; Error correcting code; Internet; Reliability; Signal to noise ratio
• ISSN: 1084-8045; 1095-8592
• DOI: 10.1016/j.jnca.2014.05.007

In this paper, we investigate the problem of packet transmissions in real-time video conferencing from a single source to multiple destinations. To overcome the burst path losses over the Internet, we propose a transmission scheme dubbed TRAding DElay for Reliability (TRADER), which appropriately spreads out the FEC (Forward Error Correction) packets׳ departures with respect to the delay constraint. The network measurements from a video conferencing system and the PlanetLab demonstrate that the transmission delays to different destinations mainly differ by 100–250ms. Motivated by this observation, we present a mathematical formulation and a systematic design to effectively leverage the delay differences for optimizing the receivers׳ video quality. The superiority of TRADER over existing transmission schemes is analyzed based on Gilbert model and continuous time Markov chain. We conduct performance evaluation via semi-physical emulations in Exata and experimental results show that TRADER outperforms the previous approaches in improving video quality in terms of PSNR (Peak Signal-to-Noise Ratio).