End-to-end optimized TCP-friendly rate control for real-time video streaming over wireless multi-hop networks

  • Authors:

  • Haiyan Luo;Song Ci;Dalei Wu;Hui Tang

  • Affiliations:

  • Department of Computer and Electronics Engineering, The University of Nebraska-Lincoln, 1110 S 67TH ST, Omaha, NE 68182, USA;Department of Computer and Electronics Engineering, The University of Nebraska-Lincoln, 1110 S 67TH ST, Omaha, NE 68182, USA;Department of Computer and Electronics Engineering, The University of Nebraska-Lincoln, 1110 S 67TH ST, Omaha, NE 68182, USA;High Performance Network Laboratory, IOA, Chinese Academy of Sciences, China

  • Venue:
  • Journal of Visual Communication and Image Representation
  • Year:
  • 2010

Quantified Score

Hi-index 0.01

Visualization

Abstract

Rate control is an important issue in video streaming applications. The most popular rate control scheme over wired networks is TCP-Friendly Rate Control (TFRC), which is designed to provide optimal transport service for unicast multimedia delivery based on the TCP Reno's throughput equation. It assumes perfect link quality, treating network congestion as the only reason for packet losses. Therefore, when used in wireless environment, it suffers significant performance degradation because of packet losses arising from time-varying link quality. Most current research focuses on enhancing the TFRC protocol itself, ignoring the tightly coupled relation between the transport layer and other network layers. In this paper, we propose a new approach to address this problem, integrating TFRC with the application layer and the physical layer to form a holistic design for real-time video streaming over wireless multi-hop networks. The proposed approach can achieve the best user-perceived video quality by jointly optimizing system parameters residing in different network layers, including real-time video coding parameters at the application layer, packet sending rate at the transport layer, and modulation and coding scheme at the physical layer. The problem is formulated and solved as to find the optimal combination of parameters to minimize the end-to-end expected video distortion constrained by a given video playback delay, or to minimize the video playback delay constrained by a given end-to-end video distortion. Experimental results have validated 2-4dB PSNR performance gain of the proposed approach in wireless multi-hop networks by using H.264/AVC and NS-2.