IBMDA: Information based misbehavior detection algorithm for VANET

• Published in: Journal of high speed networks Vol. 26; no. 3; pp. 185 - 207
• Main Authors: Singh, Dinesh, Ranvijay, Yadav, Rama Shankar
• Format: Journal Article
• Language: English
• Published: Amsterdam IOS Press BV 01.01.2020
• Subjects: Algorithms; Clusters; Flooding; High speed; Information sharing; Messages; Mobile ad hoc networks; Performance enhancement; Safety; Time lag; Topology; Traffic speed; Vehicles
• ISSN: 0926-6801; 1875-8940
• DOI: 10.3233/JHS-200638

The safety event information sharing among the vehicles in motion is the primary goal to design the vehicular ad hoc network (VANET). The shared safety event information assists vehicles to avoid road accidents and driving inconvenience. The advantages of safety event information sharing in VANET has become blunt due to the misbehavior of vehicles. The vehicle’s misbehavior like dissemination of false information, reply of bogus messages, etc., can create traffic hazards on the road and may result in the loss of property and human lives. In VANET, the detection of such misbehaving vehicles along with minimum time delay in flooding safety event information (i.e., incident delay) to others is challenging due to the high speed of vehicles. The formation of stable VANET topology is a feasible solution among many to improve the performance of misbehavior detection and reducing incident delay even with high speed of vehicles. In this paper, we propose an information based misbehavior detection algorithm (IBMDA) that effectively works in stable cluster based VANET. Our proposed IBMDA algorithm that runs on the selected cluster head vehicles is used to verify the content of received safety event messages. The identification of vehicles as malicious or non malicious depends on the result of verification at cluster heads. An illustrative example is given to explore our proposed algorithm easily and effectively. The highway scenario is considered to test the performance of our proposed IBMDA algorithm. The simulation is performed with a detailed comparative analysis using ns-3 simulator. It is observed that under the considered scenario, our proposed algorithm improves the misbehavior detection accuracy up to 6.46% and reduces average incident delay approximately up to 14.78% as compared to existing algorithms.