A Blackhole attack classification and detection system for LEACH based Wireless Sensor Network using Deep Learning approach

Wireless sensor network (WSN), are composed of sensor nodes and base station, having unique characteristics of self-organizing nature, low battery power supply, limited bandwidth support, distributed operations using open wireless medium, and multi-hop traffic forwarding. As a result WSN are vulnerable to various security attacks particularly to DoS attacks which disrupt the network operations and degrade the network performance. Specifically LEACH protocol, which is clustering algorithm used for WSN, is very exposed to the most DoS attack, which is Balckhole attack. So classification and detection system which must be compatible with the specified characteristics of LEACH based WSN and which has a capability of detecting Balckhole attack, with a low false alarm rate and high detection rate has required proposing. A novel deep learning based classification and detection system has proposed to solve those problems. For the training and testing of proposed system the dataset called Simulation-DS collected using Matlab has used. Python programming language has used for the implementation of proposed models on a Jupyter notebook of Anaconda platform. The performance metrics used were True Positive Rate (TPR), True Negative Rate (TNR), False Positive Rate (FPR), False Negative Rate (FNR), Accuracy, precision, Recall, and F1-score. Three deep learning based classification and detection models were proposed, from which CNN has outperformed the other proposed models and other related works, with high detection rates of 99.86%, and 99.99% for Blackhole attack, and Normal traffic respectively, and with low false alarm rates of 0.007%, and 0.14% for Blackhole attack, and Normal cases respectively.