Abstract :

Vehicular Ad Hoc Networks (VANETs) Play A Crucial Role In Modern Intelligent Transportation Systems By Enabling Seamless Communication Among Vehicles And Infrastructure To Enhance Road Safety And Traffic Management. With The Rapid Advancement Of Wireless Technologies Such As 5G And The Growing Adoption Of Connected And Autonomous Vehicles, VANET Environments Have Become Highly Dynamic And Data-intensive. Traditional Security Approaches, Including Rule-based Methods And Signature-based Intrusion Detection Systems (IDS), Are Increasingly Inadequate Due To Their Inability To Adapt To Evolving Attack Patterns And Large-scale Data Streams. A Key Challenge Lies In Accurately Identifying Malicious Nodes In Such Dynamic Networks, Where Frequent Topology Changes And Strict Realtime Communication Requirements Complicate Security Enforcement. Existing Systems Often Suffer From Limitations Such As Low Detection Accuracy, High False Alarm Rates, Poor Scalability, And Ineffective Handling Of Imbalanced Datasets. To Overcome These Challenges, This Study Proposes A Machine Learningdriven Framework That Integrates Data Preprocessing, Exploratory Data Analysis, And Multi-model Classification Strategies. The Framework Incorporates Models Such As Adaptive Boosting Classifier (ABC), Logistic Boosting Classifier (LogitBoost), Gradient Boosting Classifier (GBC), And A Novel Hybrid Ensemble Termed Deep Probability Boosting Classifier (DeepProbBoost), Which Combines Deep Probabilistic Neural Networks (DPNN) With Natural Gradient Boosting (NGB). Furthermore, Advanced Data Balancing Techniques, Including SMOTE-Tomek And ADASYN, Are Employed To Address Class Imbalance And Improve Model Generalization. The Proposed Hybrid Boosting Approach Demonstrates Enhanced Capability In Detecting Malicious Activities, Contributing To Scalable, Efficient, And Reliable Security Solutions For Next-generation Vehicular Networks.

Published:

10-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2244-2257

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite