Modelling and simulation frameworks for wireless sensor networks: a comparative study

Abstract

In real-time scenarios, wireless sensor network (WSN) suffers from severe constraints such as restricted power supply, short communication range, less bandwidth availability, and small memory storage. Owing to these severe constraints, the time complexity, space complexity, and cost complexity of these networks are very high. Hence, deploying a real sensor network environment for the testing of a newly designed protocol is a Herculean task. To overcome this issue, instead of doing live implementation of such networks, simulation modelling is accepted as an alternative approach. This study provides a comprehensive study about existing network simulators along with their advantages and weaknesses. This study also presented a case study for a preferable simulation methodology, i.e. MATLAB/Simulink. Firstly, this case study highlights prominent features of MATLAB/Simulink. Secondly, simulation results of implementing AODV (Adhoc on Demand Vector Protocol) in terms of simulation runtime using four simulators net simulator-2 (NS-2), NS-3, OMNet++, and MATLAB is presented. Thirdly, survey of few very well-known clustering protocols in WSNs is done. This study will prove to be beneficial for researchers of this domain as it surveys the literature over the period 2008–2020 focusing on simulation tools and simulation metrics used by researchers to simulate WSNs.