Localization Against Wormhole Attacks in Wireless Sensor Networks

Abstract

In the last decade, wireless sensor networks deployed at an accelerated pace in military, industrial, healthcare etc. fields but that were originally developed in the late 1960s and 1970s. Navigation or localization is an intuitive important because it can help in reducing the complexity of energy efficient algorithm, target tracking, routing protocol and data collection/aggregation algorithm in wireless sensor networks (WSNs). Localization has an important role in both network application domains and services; for example geographical routing. Due to self configuring nature of WSNs, it mainly deployed in a hostile environment, and hence it can be easily threatened by internal or external attack. Therefore, localization and security are essential issues in WSNs. We know that ’Wormhole Attack’ is a severe security threat where two malicious colluding sensor nodes create a virtual tunnel to WSNs. Detection and prevention of wormhole attacks in WSNs are a considerably challenging task because of its independence of MAC protocols and immunity to cryptology techniques. Around all the existing defenses have required some additional hardware requirements on the network or strong assumptions; that’s may not have perfect applicability in real environment. Our main objectives are to find the location of wormhole attackers as well as the location of sensor nodes in a wireless network system. In our schemes, we consider the three types of nodes : locators, attackers and sensors in the network and makes a conflicting set matrix on the basis of abnormal behavior of message exchanging among neighboring locators, which helps in differentiation between valid locators and dubious locators. We evaluate location estimation results through simulations in our last section.