Quality Adaptation and Resource Allocation for Scalable Video in D2D Communication Networks

Abstract

Recently fifth-generation (5G) of cellular networks appeared to enable various smart bandwidth thirsty applications. Frequency re-usability via device-to-device (D2D) communication is one of the possible ways to enhance the 5G network throughput. However, despite these technological advancements in cellular networks, users' Quality of Experience (QoE) is significantly affected by video quality adaptation according to future network conditions for video streaming. In this paper, we jointly formulate D2D pair association, video quality adaptation, and network resource allocation for scalable video streaming in a D2D communication network. The formulated problem is a mixed-integer linear programming problem and has a combinatorial nature. Thus, the formulated problem is decomposed into two subproblems: D2D pair association subproblem and resource allocation subproblem. The D2D pair association subproblem is a combinatorial optimization problem, to solve it a distributed algorithm is proposed based on two sided-matching theory. On the other hand, resource allocation subproblem is solved using conventional linear optimization techniques to get the boundary solution. Furthermore, a fully distributed quality adaptation scheme is presented using the proposed resource allocation algorithm for layered video. To validate the performance of our proposal, we perform numerical simulations for various scenarios.

Keywords

• 5G
• video quality adaptation
• video streaming
• layered video
• H.264/AVC
• SVC