Scalable video coding based video transmission in MRMC networks: A cross-layer design perspective

In this paper, we discuss the cross-layer network resource allocation scheme for video services in multi-radio multichannel (MRMC) networks. Since video services include a large amount of traffic to transmit, compared with single-radio singlechannel (SRSC) networks, MRMC networks could provide more network resource to support the large video traffic. However, MRMC networks also require more delicate cross-layer resource allocation scheme since more resource is included. In this paper, we consider the scenario that video services are encoded with scalable video coding (SVC), and each one has its own rate demand and quality level requirement. According to the various rate and quality characteristics, we jointly allocate resource in MRMC networks for these video services. We formulate a cross-layer optimization problem with joint consideration of radio assignment, channel allocation, flow routing and time scheduling according to various characteristics of services. The objective is designed as network utility maximization in order to obtain fair resource allocation. Based on the formulation, we compare the performance of the proposed scheme under SRSC networks and MRMC networks. Numerical results show that video services can be served with more resource in MRMC networks and gain higher utility than in SRSC networks. Moreover, we compare the proposed scheme which concerns various characteristics of services with conventional throughput maximization scheme. Simulation results also demonstrate that the proposed scheme performs better than conventional schemes in MRMC networks in terms of network utility.