Experiment method that follows an algorithm that rely on a repeated number of the same experiment each using different conditions (e.g. random seeds, random sampling, etc).
Published in Chapter:
Reducing Simulation Runtime in Wireless Sensor Networks: A Simulation Framework to Reduce WSN Simulation Runtime by Using Multiple Simultaneous Instances
Pedro Pinto (Instituto Politécnico de Viana do Castelo and INESC TEC, Portugal), António Alberto Pinto (CIICESI, ESTGF, Politécnico do Porto and INESC TEC, Portugal), and Manuel Ricardo (INESC TEC, Faculdade de Engenharia, Universidade do Porto, Portugal)
Copyright: © 2016
|Pages: 16
DOI: 10.4018/978-1-4666-8823-0.ch024
Abstract
Wireless Sensor Networks (WSNs) can be deployed using available hardware and software. The Contiki is an operative system compatible with a wide range of WSN hardware. A Contiki development environment named InstantContiki is also available and includes the Cooja simulator, useful to test WSN simulation scenarios prior to their deployment. Cooja can provide realistic results since it uses the full Contiki's source code and some motes can be emulated at the hardware level. However this implies extending the simulation runtime, which is heightened since the Cooja is single threaded, i.e, it makes use of a single core per instant of time, not taking advantage of the current multi-core processors. This chapter presents a framework to automate the configuration and execution of Cooja simulations. When a multi-core processor is available, this framework runs multiple simultaneous Cooja instances to reduce simulations runtime in exchange of higher CPU load and RAM usage.