MakeSense: Managing Reproducible WSNs Experiments

Wireless Sensor Networks (WSN) users often use simulation campaigns before real deployment to evaluate performance and to finetune application and network parameters. This process requires repeating the same experiments under similar conditions and to collect, parse and present data efficiently. This paper introduces MakeSense: a tool that automates this workflow and that allows reproducing simulations easily by defining the whole experiment and post-processing steps in a single JSON configuration file, easy to share and to modify. MakeSense also provides interfaces to interact with a running simulation, allowing to send external stimuli and to collect data in real time. MakeSense currently runs over the COOJA simulator, but has been built to be easily adapted to other architectures, including real testbeds.