In the event that intrusion detection could be flawlessly executed, all distant offices, data centres, and cloud properties would be safeguarded. There would be no disruption to the flow of traffic in any way. The only boundary would be metaphorically the sky.Although intrusion detection systems (IDSs) most commonly act as stand-alone devices, many contemporary and high-end firewalls incorporate IDS features. These more modern forms of protection are used in addition to traditional firewalls, which are primarily concerned with stopping known harmful traffic from accessing the network. They accomplish this by comparing the signatures or "fingerprints" in the network traffic to a larger threat database and finding complex attacks such as ransomware, data exfiltration, and network scanning. This allows them to protect the network more effectively. In many intrusion detection systems (IDSs), there is a feature known as "anomaly detection" that serves as a second layer of security and searches for deviations from regular traffic patterns that may indicate malicious intent. On the other hand, we are able to ensure the safety of intelligent transportation systems thanks to the rapid growth of mobile ad hoc networks and methods of machine learning. The authors of this work come up with an authentication technique in order to maintain the conditional privacy of vehicle users while also guaranteeing the integrity of transmitted communications. This is one of the contributions that they make. Deep learning strategies were utilized in the development of the suggested authentication solution. Researchers use the OMNeT++ application for testing the production of traffic on vehicle ad hoc network segments. Additionally, they use cutting-edge machine learning algorithms to hunt for malicious behaviour inside a specified simulated environment. With the use of this research, it was discovered that DDoS attacks may be identified with an extremely high degree of accuracy

Intrusion, Detection, Vehicular Ad hoc Networks, Deep Learning, Deep Learning