Modeling and Analysis on End-to-End Performance of Cache-enabled Networks

In this paper, we contribute to modeling and analyzing the end-to-end (E2E) packet transmission in the cache enabled networks, where the wired backhaul and the radio access network (RAN) are jointly considered. The performance of the on-peak system is first considered with modeling the two-dimensional queue process in terms of the queue length and the time. The cumulative distribution function (CDF) of the queue length and the tolerable duration of the peak time are then derived based on the compound Poisson process. On the other hand, the performance of the system in the off-peak state is further investigated. The tandem queues at the application server and the base station are proved to be independent for exponentially distributed packet sizes, and the theoretical E2E latency is elaborated. Simulation results validate the proposed model. Moreover, the analysis provides a valuable insight into the impacts of caching on both the wired backhaul and the RAN.