Enhancing intrusion detection using coati optimization algorithm with deep learning on vehicular Adhoc networks

Vehicular ad hoc networks (VANETs) role a vital play in allowing technology for future cooperative intelligent transportation systems (CITSs). Vehicles in VANETs transmit real-time data on their movement, traffic, and road conditions. Nevertheless, VANETs can vulnerable to cyberattacks, which causes serious conditions or road congestion. Intrusion detection systems (IDSs) depends on the cooperation among vehicles to identify intruders are mostly proposed security solutions for VANETs. Evolving a more efficient and effective IDS for VANETs is vital because of these networks’ unique problems and vulnerabilities. This article develops an enhanced intrusion detection using Coati Optimization Algorithm with Deep Learning (ID-COADL) technique on VANET. The major aim of the ID-COADL technique is to categorize various kinds of intrusions that exist in the VANET to accomplish security using hyperparameter tuned deep learning (DL) model. At the primary stage, the ID-COADL technique employs min–max normalization to scale the input data into uniform data. For intrusion detection, the ID-COADL technique applies deep belief network (DBN) model. Finally, the hyperparameter selection of the DBN model can be performed by the use of COA thereby enhancing the intrusion detection rate. To examine the performance of the ID-COADL system, a wide-ranging simulations were carried out. The extensive result analysis highlighted the enhanced performance of the ID-COADL technique in the detection and classification of intrusions.