An optimal trust and secure model using deep metric learning for fog‐based VANET

Abstract Vehicular ad hoc network (VANET) is a technique that enhances road safety via communication between two different vehicles. Trust establishment in VANET in between the vehicles is the most significant factor that provides secure reliability and integrity. Reliability and trust assist the ve...

Full description

Bibliographic Details
Published in:Transactions on Emerging Telecommunications Technologies
Main Authors: Tripathi, Kuldeep Narayan, Sharma, Subhash Chander
Format: Article in Journal/Newspaper
Language:English
Published: Wiley 2023
Subjects:
DML
Online Access:http://dx.doi.org/10.1002/ett.4805
https://onlinelibrary.wiley.com/doi/pdf/10.1002/ett.4805
Description
Summary:Abstract Vehicular ad hoc network (VANET) is a technique that enhances road safety via communication between two different vehicles. Trust establishment in VANET in between the vehicles is the most significant factor that provides secure reliability and integrity. Reliability and trust assist the vehicles in gathering credible data from neighboring vehicles. On the other hand, incomplete and inaccurate data gathered by the vehicles contain interrupting effect on VANET. Therefore, this article proposes a modified deep metric learning‐based chaotic Henry gas solubility optimization (MDML‐CHGSO) approach to provide a trust‐based security design. The proposed trust model is based on plausibility as well as experience in securing vehicle networks. The proposed model verifies the received data's genuineness from the approved vehicle using various processes namely, authentication, lifetime verification, experience evaluation, plausibility determination, and accurate measurement. The adoption of the modified DML‐based CHGSO approach extracts the location of the event message accurately when the location of the message is located closer to the fog nodes. In addition to this, average latency, throughput and packet delivery ratio, are the metrics evaluated for rating the performance.