Intelligent Intrusion Detection System for VANET Using Machine Learning and Deep Learning Approaches

Detecting the attacks in Vehicular Ad hoc Network (VANET) system is very important to provide more secure and reliable communication between all vehicles system. In this article, an effective Intelligent Intrusion Detection System (IDS) proposed using machine learning deep approaches such as Adaptive Neuro Fuzzy Inference (ANFIS) Convolutional Neural Networks (CNN), respectively. The existing methods focus on detecting only known VANET environment. This limitation overcome by proposing IDS soft computing techniques. method consists of Known (KIDS) Unknown (UIDS) modules, which detect both unknown attacks. KIDS module uses ANFIS classification malicious attacks, whereas UIDS a algorithm VANET. Modified LeeNET (MLNET) architecture article identify type work, DoS Botnet PortScan Brute Force are detected hybrid algorithm. obtains 96.9% Pr, 98.3% Se, 98.7% Sp, 98.6% Acc consumed 1.75 s for attack i-VANET dataset. 98.1% 98.9% 0.95 attack. 99.1% 99.2% 1.38 99.1of 97.8% 98.5% 1.29 developed methodology tested real-time CIC-IDS 2017 dataset, experimental results compared with other state-of-the-art methods.