Robust audio streaming over lossy packet-switched networks

Multimedia streaming is getting more and more important class of applications in IP networking. As real-time streaming sets contradictory requirements for network latency and error robustness, practical implementations have to trade-off between interactivity and quality experienced by the end user. In this paper a transport scheme for streaming perceptually coded audio over an unreliable packet-switched network is presented. The scheme is based on selective retransmissions and it provides improved robustness against packet loss and efficient utilization of network resources. The scheme is implemented into streaming software and it is evaluated by simulating an unreliable network environment.