Modelling and Congestion Detection of Wireless Sensor Networks: A Concurrent-based Approach using Coloured Petri Nets

Congestion in Wireless Sensor Networks (WSNs for short) causes not only packet loss and but also leads to excessive energy consumption. Therefore, congestion needs to be detected as well as controlled in order to prolong system lifetime. There are two streams to concern congestion detection including simulation-based and model-based. Following the second stream, formal modelling techniques are used for analysis of WSNs. Coloured Petri nets (CPNs for short) that combines Petri nets with programming languages is a powerful modelling technique. This paper presents a CPN-based approach for formal modelling and congestion detection of WSNs. The proposed model describes parameters and behaviours of a WSN. Then the congestion detection problem is reduced to a reachability problem on the state space of the CPN-based model. Moreover, the CPN-based model uses the hierarchical modelling capability of CPNs, including different levels of abstraction (sub-modules). This helps easily handling and extending the model. In reality, WSN components (sensors and channels) can execute a number of concurrent operations. This is called concurrency of WSNs. The CPN-based model is extended to express the concurrency, thus improving the congestion detection results.