It is not easy to make people realize that the importance of car network security is not easy.With the transition of cars to semi -autonomous driving, automobile OEMs (OEMs) are more and more concerned about car network security issues.The reason for the control of the car network is obvious. The purpose is to ensure that no one can control the vehicle except the driver (or the driver system that replaces the driver under specific and restricted conditions).

In 2021, the UNECE working group released the UN-R155, which is a network security regulations for OEM, which aims to cope with increasingly serious cyber threats.Since July 2022, the regulations have been binding on the approval of new models produced by UNECE member states.This means that car suppliers must comply with ISO 21434 to ensure that all its network security -related components meet the standard.Of course, purchasing parts that meet the network security standards do not guarantee that OEM meets the UNECE standard.However, this is an important step in this direction, so that OEM is in a more favorable position in the process of achieving this goal.This article discusses network security issues from the perspective of the image sensor used in advanced driving assistance systems (ADAS) and cabin monitoring applications.

Why do you need to ensure the safety of the image sensor

In the car, some locations should obviously implement network security, including gateway, interconnection, information and entertainment systems or any other automotive system connected through the network.However, everyone may wonder why the image sensor also needs network security.With today's attention to safety and driving assistance, the image sensor is the "eye" of the vehicle.They are used for multiple ADAS functions, such as lane deviation warnings, pedestrian detection and automatic emergency braking (AEB).They will evaluate the environment around the car and provide information input to the fusion system to make decisions.In the future, they will assist in identifying and verifying the identity of car users and monitoring users' vital signs.If the driver loses its ability, the car computer will be able to take over control.In these cases, the car image sensor must have excellent performance (high dynamic range, low -illuminarian ability, color tone power, etc.) and maintain normal work, especially in the most extreme case where the vehicle may encounter.Because the safety of cars will rely more and more to rely on image sensors, the central computer of the car needs to be authorized by authorized genuine parts to interact with it.Moreover, it is necessary to ensure that any image frame transmitted is not tampered with, and all frames are generated by genuine image sensors.In addition, the image sensor should only accept the configuration change of the automotive system instead of any other party.The following cases illustrate why the automotive industry cannot ignore the threat caused by the use of counterfeit image sensors that are vulnerable to third parties.

Threat 1: Image sensor is replaced by counterfeit parts

The AEB system relies on the image sensor behind the windshield to detect objects or pedestrians in front of the car.If the driver does not respond in time, the system can decide to apply brakes to prevent collisions.The premise of the AEB system is that its image sensor has specific characteristics (such as high dynamic range, low -illumination performance, etc.), and the system has been calibrated according to these specifications.If the original image sensor is replaced with non -genuine or counterfeit parts, it may damage the performance of the system.Although the replacement parts may look exactly the same as the original parts, its performance and characteristics may be very different.Since the AEB system is optimized for the original image sensor, the different characteristics of the replacement parts will change the performance of the system.This means that the system may only detect the objects or pedestrians in front of the car when a few meters away, so that the system does not have time to respond appropriately, which may cause tragic consequences.Replace the genuine image sensor with imitation, just let drivers with poor vision drove without glasses.

Figure 1: The consequences of using counterfeit products instead of genuine image sensors

Threat 2: Image sensor settings are modified

After calibration and programming of the vehicle system, the image sensor can be best configured to always present the scene in front of the car as much as possible.However, if someone (or object) modify the configuration of the image sensor, its performance will be affected, so that it may no longer guarantee that the car system can be correct, completely, or best to perceive the scenes faced by the car. Dust dust in the eyes of human drivers.

Figure 2: The consequences of tampering the image sensor settings

Threat 3: Image sensor is bypassed

The image sensor provides the original video data to the image processor, and then the image processor uses this data to extract key information about the front obstacles, so that the car can respond properly.For example, the image processor can detect the nearby vehicles, and then choose whether to brake or drive a car from the danger from the perspective of safety first.However, if an unauthorized party tries to tampered with the system by modifying or bypassing the image sensor, the image processor can no longer obtain the original video data that reflects the real scene.In this case, the system may no longer detect the approaching objects.On the contrary, image processing components may only receive circular images of unbroken roads without obstacles, and the consequences may be as difficult to afford as human drivers who are completely removed from the pavement.

ONSEMI's image sensor meets network security standards

Figure 3: The consequences of the image sensor are bypassed

Anson has added network security functions to its image sensor in 2018. Time is even earlier than the release of ISO 21434 network security standards.Initially, this was to meet the needs of early customers, but after gradual development, it became a valuable professional knowledge of network security.Therefore, Ansonmei's image sensor has been prepared for network security.One of the key functions is authentication, which allows them to prove to the host to prove that they are genuine. In the process, you need to use the certificate chain and pre -shared the key.Another important feature is that they can ensure the integrity of video data and prove that the video data flow between sensors and systems has not been tampered with.This integrity is provided by message verification code.In addition, the use of MAC on embedded data video lines through specific key registers can prevent sensor control and configuration data tampering.

Cyber ​​security components are the first step to realize automobile network security

Cyber ​​security compliance is a must -have condition for automotive image sensors to prevent image sensors from becoming Trojan horses that invade the complex car electronic system outside the outside world.For OEM, in order to ensure network security compliance, not only the network security control circuit is added to the image sensor, but they are critical to implement comprehensive network security compliance in ADAS and cabin monitoring systems.