HYBRID NETWORK INTRUSION DETECTION SYSTEM FOR IOT ATTACKS

A Deep Learning Dendritic Cell Algorithm (DeepDCA) is employed in an intrusion detection system (IDS) and method. The framework adopts both a Dendritic Cell Algorithm (DCA) and a Self Normalizing Neural Network (SNN). The IDS classifies interned of things (IoT) intrusion, while minimizing false alarm generation, and it automates and smooths the signal extraction phase which improves the classification performance. The IDSselects the convenient set of features from the IoT-Bot dataset, and performs their signal categorization using the SNN. Experimentation demonstrated that the IDS with DeepDCA performed well in detecting IoT attacks with a high detection rate demonstrating over 98.73% accuracy and a low false-positive rate. Also, IDS was capable of performing better classification tasks than SVM, NB, KNN and MLP classifiers.