Affiliations: Department of Computer Science, 3115 Engineering Building, Michigan State University, East Lansing, MI 48824, USA E‐mail: {smithh, mutka}@ cps.msu.edu | Department of Electrical Engineering, 2120 Engineering Building, Michigan State University, East Lansing, MI 48824, USA E‐mail: rover@ egr.msu.edu
Note: [] Corresponding author.
Abstract: This paper presents a feedback based rate control algorithm for video conferencing on the Internet. This algorithm adaptively controls the transmission rate of a video stream, which is being multicast to a heterogeneous group of receivers. The fact that the sender is transmitting a single video stream to many receivers introduces a number of issues. These issues include the variable nature of the available network resources and the fact that there is no easy way to determine the availability of these resources. In addition, even if we were able to track closely the changing network conditions, the best transmission rate for the aggregate means a trade off between high and low end receivers. The algorithm presented in this paper, called the Target Bandwidth Rate Control Algorithm, dynamically controls the output rate of the video coder by receiving and processing bandwidth levels from each receiver in the videoconference. The goal of this algorithm is to maximize the aggregate amount of video displayable at the receivers. This algorithm also allows us to limit the worst case loss experienced by the low‐end receivers. In addition to considering the available network bandwidth in the feedback information, this algorithm allows for workstation and user requirements to be considered when determining the sender’s output rate. In order to show the effectiveness of this approach, we first analyze the algorithm’s performance in a simulation environment. In this set of tests we study the effects of different configurations of receivers in terms of the number of receivers and availability of resources. Additionally, we have implemented this algorithm in the VIC video conferencing system in order to analyze its effectiveness under real network conditions.
Keywords: Video conferencing, VIC, RTP, rate control, video multicasting, adaptive applications
