ResADM‐GhostNet With Spider Wasp Optimizer: A Real‐Time DDoS Detection Framework for Vehicular Networks

VANETs are being integrated into society to ensure communication inside the networks is secured and protected. Vehicular networks are increasingly under threat from Distributed Denial of Service (DDoS) attacks that pose important threats to Intelligent Transportation System (ITS) safety and operation. This paper presents a ResADM‐GhostNet model with Spider Wasp Optimizer (SWO) for DDoS detection in vehicular networks. Data is gathered in real time (RT) from vehicular units and then preprocessed to normalize the values, remove outliers, and handle missing data. Feature extraction is performed on the traffic and spatiotemporal data patterns obtained from the network traffic and the environment where the network is deployed through packet‐level, flow‐based, and spatiotemporal features to improve the detection rate. An algorithm such as Artificial Rabbit Optimization and Kepler Optimization (ARKO) is employed to select the most relevant features. The DDoS detection model uses ResADM‐GhostNet and the SWO model to understand the dependencies, detect anomalous communication, and optimize the parameters using SWO. The proposed framework gained high efficiency in RT DDoS detection in vehicular networks compared to other models. The suggested technique attained an accuracy of 98.74%, a specificity of 98.91%, and a precision of 97.58% for 80% of the training data split.