Performance Analysis of Content Serving Through Device-to-Device Communication

This paper presents a proposed system in which content (Note: content could be any file, image, video, etc.) can deliver through device-to-device (D2D) communication. For the file delivery, a general outage probability form is derived based on the storage cache unit of the user device, the file popularity, and the signal to interference plus noise ratio (SINR). We further reduced this formula to a tractable closed-form expression. Initially, the user equipment (UE) is randomly distributed based on the Poisson point process on the cell. Then the Zipf distribution is used to model the file popularity, where the shape parameter is the most effective on the performance. The main results showed that outage probability decreases substantially with the increase of popularity shape parameter, as well as when increasing the cache storage unit of the UE. For accurate system analysis, a Monte-Carlo simulation has used to validate the theoretical results, and it showed a perfect matching between the simulation and the numerical results.