D2D Based Caching Content Placement in Wireless Cache-Enabled Networks

Abstract

In order to provide a satisfying content download service and reduce the pressure of the base station, content caching based on the D2D communications is drawing more and more attentions, which allows two requesters in close proximity to share the contents of common interest directly. However, since the storage capacity and the communication ability of the mobile nodes are limited, it is impossible for a mobile node to cache all the files and share the files with any node. Hence, one of the most important issues is the caching content placement for the mobile nodes where the files can be cached. In our work, to achieve the potential advantages of the D2D-based content caching, we consider to maximize the file download rate of the whole network by proper caching content placement. It is proved that the problem of the caching content placement is NP hard. In view of its intractability, we prove that the problem of the caching content placement can be modeled as the maximization of a monotone submodular function over one matroid and multiple knapsack constraints. Accordingly, the greedy algorithm can be utilized to obtain a suboptimal solution. It is proved that the complexity of the proposed caching content placement algorithm based on greedy algorithm is polynomial, and it yields a constant-factor approximation to the problem of the caching content placement. The simulation results show that the nodes which can cache files can perform more efficiently with our proposed caching content placement algorithm based on greedy algorithm.