Scalable audio streaming based on linear rate-MNR model

Audio streaming provides users with abundant music experiences. Scalable audio coding technology makes such experience possible with optimal quality and low cost. In this paper, adaptive audio streaming based on MPEG-4 scalable to lossless audio coding (SLS) is introduced. A model based technology is proposed for adaptive audio streaming to achieve optimal streaming quality. This scheme adaptively allocates the constraint resource to each audio frame according to the characteristics of different frames. As a result, the quality variation of streaming is reduced according to the available network condition. This scheme has low computational complexity and low memory requirement.