Content-based cross-layer design for layered video multicasting

Receiver nodes can receive the layered video streams selectively using H.264/SVC, according to the display resolution of the terminal device or congestion status of the network. The characteristics of SVC can be expressed using rate-distortion point (RDP). SVC can also be used for layered multicast. Specifically, the encoded video layers can be mapped to the proper transmission layers for layered multicast, which can be more efficient than existing single layer multicast schemes. In this paper, we propose a mechanism to select an optimal operation point from the SVC server and maximize the PSNR-RDP path. Using this, actual layered multicast is performed at the point of attachment such as BS and eNodeB in the wireless network. We also show that QoS can be improved when optimal modulations and coding schemes are used for delivering the SVC video according to the user distribution and the network condition.