Hyperparameter Tuning of XGBoost for Flooding Attack Detection in SDN-based Vehicular Ad Hoc Networks (VANETs) under Limited Resources

Software-Defined Network (SDN) based Vehicular Ad Hoc Network (VANET) infrastructure network enables centralized vehicle control. However, due to its centralized nature, SDN-based VANET is vulnerable to flooding attacks such as Distributed-Denial of Services (DDoS) or Denial of Service (DoS) attacks that can disrupt network availability and endanger traffic safety. This study aims to detect flooding attacks using the Extreme Gradient Boosting (XGBoost) algorithm with a focus on hyperparameter tuning in a limited computing environment to find optimal hyperparameter values for the model. This study uses basic Google Colab with 12 GB RAM with a total dataset of 431,371 entries. The results obtained from this study conclude that hyperparameter tuning achieves optimal performance at n_estimators = 150 and max_depth = 15, resulting in 99.97% accuracy, 99.99% precision, 99.97% recall, and 99.98% F1 score, which proves the effectiveness of the model in detecting flooding attacks. The novelty of this study lies in the application and evaluation of hyperparameter tuning on the XGBoost algorithm in a resource-constrained environment to improve attack detection in SDN-VANET.