Jensen-Shannon Divergence Based Independent Component Analysis to Detect and Prevent Black Hole Attacks in Healthcare WSN

The black hole attack is an adverse issue in Wireless Sensor Networks (WSNs). Research for detection and circumvention of the black hole attack is underway. However, the false alarm rate, amount of time required to identify the black hole attack nodes in network has not reduced. To overcome such limitations, the Jensen-Shannon Divergence Based Independent Component Analysis (JDICA) technique is proposed in this paper. This technique is introduced with the application of Jensen-Shannon Divergence estimation in Independent Component Analysis model on the contrary to existing works, in order to achieve higher black hole detection accuracy in healthcare WSN. The JDICA technique identifies the black hole attack by analyzing the physiological data gathered from biomedical sensors. The proposed JDICA technique carries out attack detection based on sensor nodes behaviors such as energy, trust and cooperative count. It determines the dependence among the nodes, based on the independent probability distribution functions and mutual probability function by using the Jensen-Shannon Divergence. The divergence result enables JDICA technique to detect black hole attacks with greater accuracy, and helps to quarantine the malicious node from the network by broadcasting the isolation message to all sensor nodes in the network. Hence, JDICA technique enhances the detection of black hole attack nodes as compared to state-of-the-art works, thereby increasing the packet delivery ratio and reducing delay. The JDICA technique simulation is done considering the metrics such as detection rate, detection time, false alarm rate, and packet delivery ratio with respect to a varied number of sensor nodes and data packets. Simulation results makes it apparent that the JDICA technique improves the detection rate and minimizes the detection time of the black hole attack when compared to state-of-the-art works.