Bernoulli Vacation Scheduling (BVS) Model for Energy efficient intrusion detection in WSN

Abstract

Cutting-edge wireless sensor networks, security is the most important challenge that has to be considered. Most of the applications of WSN are vulnerable to attacks, as they are deployed in any kind of environment. So, the intrusion detection system is the second line of defense for WSN. The intruder should be detected in WSN more efficiently. We introduce the Bernoulli scheduling rule such that several trials are performed to get the detection of an intruder to be successful once. Since the sensor nodes are always active for detecting an intruder, the energy consumption becomes high. To reduce the consumption of energy, sleep/wakeup, also called small interval vacations are used for the sensors which are almost considered to be like the mechanism of fluid queues. In our proposed work, the vacations are introduced so that the energy used is minimized whenever the sensor is not required to be active. Thus, in our proposed work, since WSN is vulnerable to many harmful attacks, the Bernoulli scheduling rule is applied and the sensor nodes have a vacation state whenever the sensor need not be active, to save the energy consumed. The probability of detecting an intruder is calculated and performance is compared. We have applied a fluid queuing model for the vacations of small intervals in the sensor nodes to minimize the energy consumption and the obtained result is compared with the energy consumed when the vacations are not applied. Thus, the results are simulated for both energy consumed and the number of success rates of detection of an intruder, most importantly the comparison is analyzed with the past work of wsn.