Scalable Video Multicast for Multi-Cell Cellular Wireless Networks

Hung-Bin Chang, Izhak Rubin, Ofer Hadar
2015 Journal of Communications  
Abstract-We consider adaptive-rate multicasting of video streams compressed by using Scalable Video Coding (SVC) over cellular wireless networks. Base station nodes coordinate the scheduling of their downlink transmissions to mitigate intercell interference; as well as, at times, to direct the multicast transmission of the same video streams to a group of clients, as implemented in a Multicast-Broadcast Single-Frequency Network (MBSFN) of LTE system operation. To enhance performance, we study
more » ... formance, we study the optimal configuration of a Fractional Frequency Reuse (FFR) scheme. We classify cell users into two groups, based on their experienced SINR (signal-tointerference-plus-noise ratio) levels. Lower-SINR group users will be offered streams at a lower video quality level, while higher-SINR group members will be provided streams at a higher video quality level. For this purpose, through the proper joint setting of the channel encoding rate, video encoding rate and the scheduling schemes, we calculate the minimum bandwidth required per stream. We demonstrate the optimal design of such a system by considering a utility function that includes a metric expressing the user's willing-to-pay when the reception is of a sufficient high video-quality. We show that, under high cell coverage requirement, the proposed optimally configured FFR scheduling schemes offer a significantly enhanced performance behavior. We also show that such an MBSFN scheme that involves coordinated cooperation among 3 cells, combined with the optimal FFR scheme, outperforms a system that employs a regular MBSFN operation that involves cooperative transmissions among 7 and 19 cells. Index Terms-MBSFN, fractional frequency reuse, cellular wireless networks, scalable video coding, quality of experience, willing to pay model, multicast, cell coverage
doi:10.12720/jcm.10.9.715-727 fatcat:fuaicp3vn5g6dmxeapaoyjehj4