Robust low-latency voice and video communication over best-effort networks.

摘要

The quality of service limitation of today's best-effort networks poses a major challenge for low-latency multimedia communication. Excessive delay, packet loss, variations in throughput, and high delay jitter all impair the performance of the communication. In this work, these challenges are addressed at transport and application layers of real-time and on-demand streaming media systems.; On the client side, passive schemes including adaptive playout scheduling and low-latency loss-concealment have shown to significantly improve the trade-off between buffering delay and packet loss for real-time voice communication. The playout schedule of the media packet is adaptively adjusted in a highly dynamic way and the proposed underlying packet scaling technique based on time-scale modification works elegantly and preserves good sound quality.; At the transport layer, the communication performance is further improved by exploiting diversity of multiple transmission channels, where the source media are coded into multiple complementary streams that are sent over independent network paths. Experiments demonstrate further gains of reduced latency and distortion, resulting from path diversity.; In order to combat network losses for real-time and on-demand video communication, which exhibits stronger dependency across packets, a network-adaptive coding scheme is employed to dynamically manage the packet dependency using optimal reference picture selection. The selection of the reference is achieved within a rate-distortion optimization framework and is adapted to the varying network conditions.; For network-adaptive streaming of pre-encoded media, the potential mismatch error during bitstream assembly is avoided by using a layered coding structure. Based on an accurate loss-distortion model introduced in this work, a prescient scheme that optimizes the dependency of a group of packets is also proposed to achieve global optimality as well as improved rate-distortion performance. With the improved trade-off between compression efficiency and error resilience, the proposed system does not require retransmission of lost packets, which makes low-latency communication possible.; These solutions provided at the receiving client, the transport layer, and source coding significantly improve the perceptual quality and reduce the latency for media communication over best-effort networks, without any requirement to modify the current or future network infrastructure.