Modeling and analyzing cascading dynamics of the clustered wireless sensor network

Although some progress has been made in studying the invulnerability of wireless sensor networks (WSNs) against cascading failures, they only consider the impact of network topology on network invulnerability, and do not consider the impact of routing protocols, which makes these studies unable to correctly reflect the traffic characteristics of WSNs. In this work, we construct a cascading model for clustered WSN and proposed a congestion-aware routing recovery mechanism. In this model, the load function is defined on each node according to the real-time number of data packets, and the overload function is defined on the basis of the congestion state of each node. The overloaded node can recover after a certain time delay, instead of being deleted permanently from the network. We analyzed the impact of key parameters and evaluate the performance of several typical clustered routing protocols from the perspective of cascading invulnerability. The simulation results show that the network invulnerability is positively correlated with overload tolerance coefficient and negatively correlated with congestion tolerance coefficient, and the extension of recovery time can aggravate the fluctuation of the cascading process. In addition, balancing the cluster size and putting the sink node near the center of the deployment area can help the network reduce the risks of cascading failures. By introducing routing recovery mechanism into existing clustered routing protocols, the network is able to recover in a few steps.