Enhancing VANET Security Through ANFIS‐Based Intrusion Detection and Element Perturbation
International Journal of Communication Systems · 10.01.2026 · 39 (1)
DOI: 10.1002/dac.70305 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Communication Systems, online erschienen am 18.11.2025. Verfasst von G. Mahalakshmi und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 28 Literaturangaben.
Abstract (Originalsprache)
Vehicular ad‐hoc networks (VANETs) serve an important role in enabling intelligent mobility by allowing vehicle‐to‐vehicle (V2V) communication. However, due to their open and dynamic character, they are extremely vulnerable to a variety of internal and external threats, providing substantial hurdles to intrusion detection and data privacy. To address these problems, this study introduces a novel element‐based intrusion detection model (EIDM) that combines the adaptive neuro‐fuzzy inference system (ANFIS) with element perturbation computations and differential privacy approaches. The proposed system assures secure and privacy‐preserving data sharing between cars, reduces security discrepancies between nodes, and allows for real‐time alert broadcast to improve traffic safety and situational awareness. EIDM, unlike classic approaches, combines local training at vehicle nodes with collaborative detection, providing scalability and responsiveness while maintaining data confidentiality. The proposed system combines ANFIS with element perturbation and differential privacy in a unique way, allowing for precise and real‐time intrusion detection while maintaining data confidentiality, in contrast to traditional IDS approaches that either compromise privacy or require a lot of computation. Because it tackles the crucial trade‐off between scalability, privacy protection, and detection accuracy, this hybrid approach is ideal for dynamic VANET systems. The model is validated using large NS2 simulations and outperforms contemporary models such as deep belief networks (DBN) and neural networks (NN). It has a sensitivity of 96.3%, specificity of 93.8%, precision of 98.6%, detection accuracy of 94.1%, and packet delivery ratio of 93.9%. These findings confirm EIDM as a strong, scalable, and efficient approach for improving VANET security and dependability.
Zitieren
APA
Mahalakshmi, G., Sumathi, G., Prakash, M., & Solainayagi, P. (2026). Enhancing VANET Security Through ANFIS‐Based Intrusion Detection and Element Perturbation. International Journal of Communication Systems, 39(1). https://doi.org/10.1002/dac.70305
Vancouver
Mahalakshmi G, Sumathi G, Prakash M, Solainayagi P. Enhancing VANET Security Through ANFIS‐Based Intrusion Detection and Element Perturbation. International Journal of Communication Systems. 2026;39(1). doi:10.1002/dac.70305
BibTeX
@article{Mahalakshmi2026,
author = {Mahalakshmi, G. and Sumathi, G. and Prakash, M. and Solainayagi, P.},
title = {Enhancing VANET Security Through ANFIS‐Based Intrusion Detection and Element Perturbation},
journal = {International Journal of Communication Systems},
year = {2026},
volume = {39},
number = {1},
doi = {10.1002/dac.70305}
}
EUVIMED