ML-Based Approach to Detect DDoS Attack in V2I Communication Under SDN Architecture

The need for Internet-based services is increasing at a tremendous pace in smart cities. The driver and occupants of the vehicle access Internet, and different intelligent transportation system (ITS) related services such as real-time traffic information, parking space availability, downloading the map, etc., in a vehicle to infrastructure communication (V2I) mode. In a highly dynamic network environment like vehicular network, software-defined networking (SDN) promises to be an ideal solution. However, it also opens doors for various distributed denial of service (DDoS) attacks. An attacker can easily flood short-lived spoofed flows and exhaust network resources. This motivates us to find a solution to detect the attacks in a V2I communication under SDN. In this paper, we propose a machine learning (ML) based DDoS attack detection. The proposed system uses various ML schemes, and few of them found to be accurate with a high detection rate and a relatively low false alarm rate.