EasiSim: A Scalable Simulation Platform for Wireless Sensor Networks

Abstract

Traditional simulators cannot meet the requirement of modeling large scale networks due to their deficiency in scalability. In this paper, we present a new simulator, namely EasiSim, for sensor networks on a large scale. EasiSim is featured by a structure-based modeling method and a hierarchical organization of the relevant functional components, including nodes, topology and scenario. The nodes are organized into a three-dimensional sorted list (3D list), which enables the node to process all the concurrent events in one batch, and therefore the running time may be reduced by an order of magnitude. Integrated with the other two upper layer components, which are topology and scenario, the proposed node organization method based on the 3D list makes the simulator not only scalable but also extensible. Moreover, we propose a visualization scheme based on a client/server model which separates the graphical user interface (GUI) from the simulation engine, and therefore the scalability of the simulator will not be decreased. The performance of EasiSim is evaluated through extensive simulations and compared with ns-2 in terms of real running time and memory usage. The results show that EasiSim takes less time and less memory than ns-2 to complete simulations with the same number of nodes during a same configured simulation time.