A Mass-Processing Simulation Framework for Resource Management in Dense 5G-IoT Scenarios

Abstract

Because of the increment in network scale and test expenditure, simulators gradually become main tools for research on key problems of wireless networking, such as radio resource management (RRM) techniques. However, existing simulators are generally event-driven, causing unacceptably large simulation time owing to the tremendous number of events handled during a simulation. In this article, a mass-processing framework for RRM simulations is proposed for the scenarios with a massive amount of terminals of Internet of Things accessing 5G communication systems, which divides the time axis into RRM periods and each period into a number of mini-slots. Transmissions within the coverage of each access point are arranged into mini-slots based on the simulated RRM schemes, and mini-slots are almost fully occupied in dense scenarios. Because the sizes of matrices during this process are only decided by the fixed number of mini-slots in a period, the time expended for performance calculation is not affected by the number of terminals or packets. Therefore, by avoiding the event-driven process, the proposal can simulate dense scenarios in a quite limited time. By comparing with a classical event-driven simulator, NS2, we show the significant merits of our proposal on low time and memory costs.