A Key Level Selection Within Hash Chains for the Efficient Energy Consumption in WSNS

Abstract

A wireless sensor network is composed of a base station (BS) and numerous sensor nodes. The sensor nodes lack security because they operate in an open environment, such as the military. In particular, a false report injection attack captures and compromises sensor nodes. The attack then causes the compromised nodes to generate forward false reports. Owing to the false report injection attack, not only does the sensor network have a false alarm, but its limited energy i s also drained. In order to defend the false report injection attack, over the past few years, several studies have been made looking for a solution to the attack. Ye et al. studied statistical en-route filtering (SEF). SEF is a method of stochastically ve rifying event reports in the en-route filtering phase. SEF can filter many false reports early using verification of intermediate nodes. However, because the number of keys in a sensor node is fixed by the system, the sensor network cannot control the even t report verification probability depending on the circumstances of the network. Therefore, it is difficult to efficiently consume energy of the sensor network. In order to solve the problem, we propose a method which controls the event report verification probability by using a key sequence level of an event report. In the proposed method, when an intermediate node receives an event report, the node verifies the event report by comparing a key sequence level of the report and its key sequence level. Elements determining the key sequence level include the density ofneighbour nodes in the sensing range of a center of stimulus (CoS), the number of hops from the CoS to the BS, and the average of the key sequence level of intermediate nodes in each path. We sim ulated the proposed method and the SEF method to evaluate the performance in terms of energy efficiency and security. In the simulation results, the proposed method consumed an average of 7.9% less energy of the sensor nodes compared to SEF method. The number of false reports arriving at the BS of the proposed method was also less, by an average of 6.4, compared to the SEF method. Through the results, we can see that when the number of false report is large in the sensor network, the proposed method is moreenergy-efficient and secure than the SEF method.