Malicious Data Injection Detection and Prediction in Wireless Sensor Network using Optimized Swarm Intelligence

Abstract

Wireless sensor networks (WSNs) are vulnerable, that is, they can be physically or remotely exposed to destructive effects. In other words, security is critical in many wireless sensor network applications. Sensor readings are used to locate events such as fires, floods, etc. Because wireless sensor networks are vulnerable, there is a need to protect the network by detecting the introduction of false data. We have developed an algorithm to identify and fix malicious data in the network. The developed optimized swarm intelligence algorithm is applied to different datasets to analyze its performance. A simulator is used to test the algorithm. The analysis and simulation results show the detection and correction of malicious data in wireless sensor networks. In this paper we test simulation of data injection attack in WSN and reach the accuracy of 95% using Matlab.