Machine Learning-Based Malicious Vehicle Detection for Security Threats and Attacks in Vehicle Ad-hoc Network (VANET) Communications

TL;DR

This work addresses the security risk of blackhole attacks in Vehicle Ad-hoc Network (VANET) communications by developing a machine learning-based detection approach. It builds a labeled dataset from NS-3 simulations using AODV routing, extracts a compact, discriminative feature set (outgoing/destination addresses and ports), and evaluates six classifiers (GB, RF, SVM, KNN, GNB, LR) for early malicious-node detection. Gradient Boosting delivers the best overall performance with a reported accuracy of $94.82\%$, high sensitivity, and strong ROC AUC, demonstrating the feasibility of ML-powered VANET defense. The findings highlight the potential of ML to enhance VANET security by promptly identifying and mitigating blackhole attacks, with future work focusing on feature refinement and real-time implementation.

Abstract

With the rapid growth of Vehicle Ad-hoc Network (VANET) as a promising technology for efficient and reliable communication among vehicles and infrastructure, the security and integrity of VANET communications has become a critical concern. One of the significant threats to VANET is the presence of blackhole attacks, where malicious nodes disrupt the network's functionality and compromise data confidentiality, integrity, and availability. In this paper, we propose a machine learning-based approach for blackhole detection in VANET. To achieve this task, we first create a comprehensive dataset comprising normal and malicious traffic flows. Afterward, we study and define a promising set of features to discriminate the blackhole attacks. Finally, we evaluate various machine learning algorithms, including Gradient Boosting, Random Forest, Support Vector Machines, k-Nearest Neighbors, Gaussian Naive Bayes, and Logistic Regression. Experimental results demonstrate the effectiveness of these algorithms in distinguishing between normal and malicious nodes. Our findings also highlight the potential of machine learning based approach in enhancing the security of VANET by detecting and mitigating blackhole attacks.

Figures (4)

• Figure 1: Illustration of a Blackhole Attack in VANET Communications.
• Figure 2: Malicious Vehicle Detection Process Pipeline.
• Figure 3: Performance comparison of 6 different machine learning algorithms. GB algorithm achieved the highest performance (accuracy 94.81% and sensitivity 97.88%). Legend: Accuracy(Acc), F1-score (F1), Negative Predictive Value (NPV), Positive Predictive Value (PPV), Sensitivity (Sen).
• Figure 4: ROC AUC Score Comparision of 6 Different Machine Learning Algorithms.