Performance improved security system in wireless sensor networks using classifier

Abstract

Recently Wireless Sensor Networks WSNs have received much attention as a means for collecting and utilizing data from real world The number of WSN applications has been increasing widely and the application range is expected to spread WSN is a network composed of a large number of sensor nodes with limited radio capabilities and one or few sinks that collect data from sensor nodes Residual nodes are the malicious nodes in WSN which degrades the performance of the WSN The malicious node detection is a multifaceted crisis in WSN due to its analogous distinctiveness with other nodes in with other nodes in WSN. Trust model is developed to extract the features of the nodes in the sensing area Further feed forward back propagation algorithm is applied on the extracted features to detect the newlinemalicious nodes in WSN The detected malicious node is held in a blackhole list and all other nodes participating in the routing of the message are warned to stop communicating with the malicious node in that list The malicious node detection system is analyzed in terms of Sensitivity Specificity Accuracy Packet Delivery Ratio Latency and Detection Rate with State of art methods newline newline