Misbehavior detection system in VANETs using local traffic density

Abstract

In this paper we explore a novel approach for misbehavior detection in Vehicular Ad-Hoc Networks (VANETs) using local traffic density. The approach is based on measuring local traffic density from two independent sensors and representing it as evidence for certain traffic situation. Dempster rule of combination is used for fusing together multiple pieces of evidence from reliable and unreliable sensors to detect the misbehavior. The approach is particularly suited to detect illusion attacks, which is still a challenge for vehicular communication. We motivate and discuss the approach and demonstrate its potential by an example scenario considering illusion attack.