A DDoS Attack Detection Method Based on LSTM Neural Network in The Internet of Vehicles
Yuexin Zhang, Yiyang Liu, Yiying Zhang, Longzhe Han, Jia Xi Zhao, Yannian Wu · 2021
With the continuous development and application of 5G wireless communication technology and smart car technology, the Internet of Vehicles has attracted more and more attention. Due to the application requirements of high bandwidth and low latency in the Internet of Vehicles, Mobile Edge Computing (MEC) is introduced into the in-vehicle network. However, when mobile edge computing nodes are deployed near the edge of the Internet of Vehicles to bring users a good experience, more and more attacks on the Internet of Vehicles will follow. Distributed denial of service attacks are violent and direct, but they cause huge damage to the car networking system. Aiming at the problem that the existing DDoS attack detection methods are not suitable for the Internet of Vehicles environment and have hysteresis problems. This paper proposes a detection method based on LSTM neural network in the mobile edge computing environment of the Internet of Vehicles. This method can detect the current network traffic according to the LSTM prediction model, compare the current data with the prediction data of the LSTM detection model, and judge whether the car networking system is under DDOS attack according to the threshold value; in addition, the method can also continuously learn historical information and build local knowledge The database is continuously detected according to the attack signature database of the knowledge base.