AN ADAPTIVE STREAMING SYSTEM FOR MPEG-4 SCALABLE TO LOSSLESS AUDIO

In this paper, we propose an adaptive streaming system for MPEG-4 Scalable to Lossless (SLS) encoded audio. In the proposed system, the fine grain scalable (FGS) feature of SLS is utilized to achieve optimal audio streaming quality through network with limited and possibly time-varying network bandwidth. To this end, the streaming system selects an optimal target quality according to the available network resources and rate-quality relationship of SLS encoded frames, and the SLS frames are then truncated according to the target quality before they are transmitted to the client. The proposed system adapts to both the network conditions and the rate-distortion characters of the SLS bit-stream so that the quality of the streaming audio is optimized without over-utilizing the network resources.