Optimal Control Model for Attack of Worms in Wireless Sensor Network

Wireless Sensor Network (WSN) is composed of large number of sensor nodes which communicate with each other through wireless medium. Such networks are prone to the attack of self-replicating worms known as malware used to bring different form of attacks on the WSN during the communication of the sensor nodes. Efficient countermeasures to anticipate malware hazards are to be developed and understand the threats they pose before they emerge in the hands of the attackers. In this paper, we propose security mechanisms using three different epidemic models with proper countermeasure which depends on the incidence of worm infections. A special crashed compartment is introduced that includes the nodes which crash out due to induced infection or reason other than the attack. The crashed class is expected to provide a new view-point which can help in optimizing the level of infection and of corresponding recovery of sensor nodes in the wireless sensor network. The past and present information of the worms represented in the form of Information variable and a Control function to minimize the attack of malware in WSN is taken into account in the different epidemic models developed. Further threshold values have been obtained that determine whether the worms completely die out or not. The stability of the worms-free and endemic equilibrium of the epidemic models is well discussed and it is also shown that the models may undergo a forward bifurcation. To validate the models extensive simulation is carried out. The results may help in developing a proper secure mechanism in WSN.