Deep learning intrusion detection system for mobile ad hoc networks against flooding attacks

Abstract

Mobile ad hoc networks (MANETs) are infrastructure-less, dynamic wireless networks and self-configuring, in which the nodes are resource constrained. With the exponential evolution of the paradigm of smart homes, smart cities, smart logistics, internet of things (IoT) and internet of vehicle (IoV), MANETs and their networks family, such as flying ad-hoc networks (FANETs), vehicular ad-hoc networks (VANETs), and wireless sensor network (WSN), are the backbone of the whole networks. Because of their multitude use, MANETs are vulnerable to various attacks, so intrusion detection systems (IDS) are used in MANETs to keep an eye on activities in order to spot any intrusions into networks. In this paper, we propose a knowledge-based intrusion detection system (KBIDS) to secure MANETs from two classes of distributed denial of service (DDoS) attacks, which are UDP/data and SYN flooding attacks. We use the approach of deep learning exactly deep neural network (DNN) with CICDDoS2019 dataset. Simulation results obtained show that the proposed architecture model can attain very interesting and encouraging performance and results (Accuracy, Precision, Recall and F1-score).