Design and Performance Analysis of Cache-Enabled Multicast in UAV-Assisted Cellular Networks

Abstract

The temporary events are generally gathered around many users interested in the same content. An unmanned aerial vehicle (UAV) with caching and multicasting is attractive for such scenarios with high traffic demands, since the multicasting allows concurrently serving users, and the caching can alleviate the burden on backhaul links. Hence, in this work, we investigate a cellular network assisted by caching-and-multicasting-empowered UAVs, where UAVs multicast the files from their caching storage or base stations via wireless backhaul links. Particularly, a popularity-aware (PA) file selection and multicast scheme is proposed, where the files to be multicasted are determined by the instantaneous requested popularity, the maximum number of allowable fetched files and the performance on wireless backhaul links. By leveraging stochastic geometry, we obtain the approximated yet accurate result for the average number of successfully multicasted users. Our results confirm the effectiveness of the PA scheme, i.e., achieving a larger average number of successfully multicasted users compared to the random scheme for most cases. Moreover, the results suggest that a relatively smaller number of multicast channels or reducing the maximum number of allowable fetched files can benefit network performance in the case of the high signal-to-interference ratio threshold on the backhaul links.