QoE-Aware Optimization for SVC-Based Adaptive Streaming in D2D Communications

As the growing demand for the large-scale and diverse services, Device-to-Device (D2D) communications have been considered as one of the most promising methods in 5G to improve the capacity of the service network and the quality of experience (QoE) for mobile users, especially those located in specific areas, asking for the same kind of multimedia services. However, it is still tough for service providers to manage these devices in a short time because different users have different QoE requirements and the arrival and departure of each user are random, which severely affect the continuity of service for other neighbor users. In this paper, we propose a fast D2D Multimedia Transmission Scheme (named DMaster) based on local search to overcome these drawbacks and solve the D2D service initialization problem, which is NP-Complete. DMaster organizes D2D communications as a spanning tree transmitting SVC-based Adaptive Streaming (SVC-AS). Furthermore, we prove that the running time of DMaster utilizing disjoint-set is O(vertical bar L parallel to D vertical bar alpha(vertical bar L vertical bar; vertical bar D vertical bar)), where L is the set of potential D2D links, D is the set of devices, and alpha is the inverse Ackerman function. Extensive simulations show that the proposed scheme guarantees QoE, fairness, and continuity of service and supports a large number of users. To the best of our knowledge, it is the first work focusing on the SVC-based fast service scheme in D2D communications.