Publication

Abstract : VANET is a highly dynamic network, where the vehicles frequently move around various locations. Due to its rapidly changing topology and unreliable security infrastructure, the routing protocols in VANET are vulnerable to several attacks such as DDoS attacks, blackhole attacks, and wormhole attacks. This paper focuses on a cooperative blackhole attack where several malicious nodes collaborate to execute the attack. The attacker nodes drop all packets they receive. We present a security technique to detect the cooperative blackhole attackers by analyzing the dropped packets at each node. Using linear regression to determine the packet drop threshold helps our proposal to improve its accuracy further. The simulation results show that our proposed system provides a high detection accuracy of 99.78% and false positives limited to 0.025%.