Fuzzy aggregated rate controller for H.264/ scalable video coding with dependent layer quantization

This paper presented a fuzzy rate control algorithm for scalable extension of H.264/AVC (H.264/SVC video coding standard). The algorithm operates on the group of picture (GOP) level of scalable layers. It aggregates the bit streams of all scalable layers in a virtual buffer and adjusts the QP of the base layer according to a feedback signal from the buffer status. Considering the dependency of SVC layers, the QPs of the enhancement layers are computed with respect to the derived QP of the base layer. This method guarantees a constant relative higher quality for each enhancement layer. From the quality consistency point of view, as well as delay constraint, the algorithm is optimized to be used for a wide range of variable bit rate (VBR) applications. Simulation results show that the proposed fuzzy controller provides much higher average quality in comparison with the constant QP (CQP) method while, the buffer constraint is completely obeyed.