Clone Attack Detection using Random Forest and Multi Objective Cuckoo Search Classification

Abstract

Intrusion Detection Systems (IDSs) have played a significant responsibility in recognizing and preventing security attacks in Wireless Sensor Networks (WSNs). Modelling of IDS should be done in WSN to guarantee dependability and security of WSN services. In this work, an approach is designed to detect various kinds of clone attack in WSN. In specific, an Adaptive random Forest based Multi-objective Cuckoo Search algorithm (RF-MOCS) is designed to identify the source of clone attack using KDD cup dataset. The proposed model provides significant performance in terms of accuracy, sensitivity, specificity, F-measure respectively. The proposed design shows better trade off when compared to existing techniques like ANN, Naive bayes, SVM.