Search

Modern mobile networks in the era of 5G have ambitious performance goals. One envisioned goal is to have communication that is ultra-reliable and ultra-low latency (uRLLC). To ensure reliability, Automatic Repeat Request (ARQ) processes are usually deployed, sometimes combined with forward error correction. As an evolution of the Long Term Evolution (LTE) system, 5G may be foreseen to use similar methods, and so we assume LTE's Hybrid ARQ on the Medium Access Control (MAC) layer, and a Selective Repeat ARQ on the Radio Link Control (RLC) layer. Simulation of LTE systems is a time-consuming endeavor due to its complexity, and when latency and reliability are the key performance indicators, it may be preferable to isolate the ARQ processes. In this manner, this paper presents a mathematical model of the two stacked ARQ processes that are used in today's LTE system. It is based on the method of Signal Flow Graphs (SFGs) and allows the analysis of either ARQ process in isolation, and of both stacked on top of each other. Analytic results are shown for a realistic set of parameters, and in the spirit of Open Science, the MATLAB implementation files are made available in the conclusion to interested researchers, so that verification and an adaption to similar ARQ processes is easily made.

Next generation mobile systems aim to meet the strict demands of an exponentially increasing number of users. Particularly device-to-device (D2D) communication has attracted a lot of attention during the last years. D2D communication can ofïoad cellular trafïc off the base station. It may reuse radio frequency resources simultaneously used by cellular communication for D2D transmissions, thereby increasing spectral efïciency and system capacity. However, reusing frequency resources can lead to in-cell interference. The expected interference must therefore be actively mitigated as part of the scheduling process. We investigate a novel joint radio resource scheduling and allocation algorithm for D2D communication that makes use of two methods from the ïeld of game theory. It aims to meet the respective userâ™s Quality of Service requirements and strives for fairness between users by playing a Transferable Utility game. At the same time it actively mitigates in-cell interference from frequency resource reuse and determines an optimal transmission power through Stackelberg gameplays. It is shown through simulation that system performance increases if resource allocation and reuse is done in a coordinated manner. The proposed scheduler scales well due to its polynomial computational complexity and outperforms baseline schedulers and the single-game approach in several simulated scenarios.