Social-aware collaborative caching for D2D content sharing

With the rapid growth of wireless content demands, Device-to-Device (D2D) content sharing technology is proposed to effectively alleviate the pressure of base stations and improve the quality of service of users. However, due to the limited storage capacity of devices, the various content demands are difficult to be satisfied. Hence, caching schemes are needed. In particular, the collaborative caching, which can increase the utilization ratio of the storage capacity, attracts much attention. Moreover, we introduce the social popularity to improve the availability of preset contents, and then, we propose the social-related download rate by combining the physical and social information. Guided by this, we model the social-aware collaborative caching problem in a D2D content sharing scenario by maximizing the sum of social-related download rate over the constraint of limited storage capacity. Due to its intractability, it is computationally reduced to the maximization of a monotone submodular function, subject to a partition matroid constraint. Subsequently, the social-aware collaborative caching algorithm based on greedy algorithm is designed to achieve a suboptimal solution within a factor 1/2 approximation guarantee and polynomial-complexity.