A buffer dynamic stabilizer for low-latency adaptive video streaming

Abstract:

HTTP-based adaptive streaming is today's predominating video delivery model over Internet. However, state-of-the-art bitrate adaptation algorithms are not suitable for low-latency adaptive video streaming due to a distinct lack of buffer dynamic stabilization at the client. A small buffer size lacks resilience to the buffer fluctuation imposed by the mismatch between the network throughput and the average bitrate of the selected video chunk. This paper introduces a bitrate adaptation algorithm that minimizes buffer deviation from the desired level despite a buffering delay as small as a chunk-duration. We demonstrate the performance of our algorithm with respect to the quality-of-experience (QoE) using two evaluation frameworks and compare it against two well-known streaming algorithms, FESTIVE and DASH VLC plugin. The preliminary results show that our algorithm achieves at least 11% higher QoE in the simulation framework and 4% in the emulation framework.