Smart vehicle monitoring for detecting anomalies on CAN bus

Abstract

Modern vehicles are embedded with advanced electric devices known as Electronic Control Unit (ECU). CAN protocol is used in transmitting/receiving messages between the ECUs. CAN bus is a well-known and broadly used protocol in the automotive industry and is evaluated as one of the best standards for vehicular networks. Unfortunately, the CAN bus protocol is vulnerable to frequent attacks due to the lack of a message authentication capability in the protocol. Unprotected CAN communications make the vehicles accessible to a variety of harmful wired/wireless attacks. This makes easier for hackers to break the network and carry out malicious activities. In the proposed system, CAN bus protocol is implemented to monitor the parameters like Vehicle Speed, Seatbelt status, Temperature, humidity, Altitude of the vehicle remotely. This system will be able to overcome the security drawbacks in CAN bus protocol which detects and ignores any irregular or unusual traffic activities such as fuzzy, Denial of Service (DoS), and impersonation attacks using the anomaly detection method.