CAT: Consensus-assisted trust estimation of MDS-equipped collaborators in vehicular ad-hoc network

• Published in: Vehicular Communications Vol. 2; no. 3; pp. 150 - 157
• Main Authors: Raghu Vamsi Krishna, T., Barnwal, Rajesh P., Ghosh, Soumya K.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.07.2015
• Subjects: Misbehavior; Security; Trust; VANET; VANET; Security; Misbehavior; Trust
• DOI: 10.1016/j.vehcom.2015.06.001
• ISSN: 2214-2096

The trust establishment in highly mobile ad-hoc network is a challenging task. Vehicular ad-hoc network (VANET) poses trust estimation problem in dynamic network environment. Presence of misbehavior detection schemes (MDS), in the vehicular node, safeguards them up to a certain extent from misbehaving vehicles but existing MDS algorithms have their own limitations in terms of capability and accuracy. In this scenario, vehicles are supposed to collaborate among themselves for getting better assessment of the misbehaving peers. However, the trustworthiness of collaborators is always a questionable factor subject to verification. Trust estimation of the collaborating nodes is essentially required to provide supplementary information for more accurate decision about the adversaries. In this paper, algorithms have been proposed to estimate trustworthiness of the MDS-equipped collaborating nodes with the help of consensus mechanism. After comparison of different proposed trust-estimation algorithms in simulated environment, results reveal that the unsupervised k-means based consensus clustering algorithm is an effective solution for precise estimation of the trustworthiness of the collaborating node.