Extracting Markov chain models from protocol execution traces for end to end delay evaluation in wireless sensor networks

Many WSN industrial applications impose requirements in terms of end to end delay. However, the end to end delay estimation in WSNs is not a simple task because of the high dynamic of networks, the use of duty-cycled MAC protocols as well as the impact of the routing protocols. Markov-based modelling is an interesting approach to deal with this problem aiming to provide an analytical model useful for understanding protocol's behavior and to estimate the end to end delay, among other performance parameters. However, existing Markov-based analytic models abstract the reality simplifying the analysis and thus resulting models are not accurate enough for estimating the end to end delay. Furthermore, establishing an accurate Markov model using classic approaches is very difficult considering the highly dynamic behavior of the sensor nodes. In this paper, we propose a novel approach to obtain the Markov chain model of sensor nodes by means of Process Mining techniques through the code execution trace. End to end delay is then computed based on this Markov chain. Experimentations were done using IoT-LAB testbed platform. Comparisons in terms of delay are presented for two different metrics of the RPL protocol (hop count and ETX).