MadRadar: Engineers Develop a Hack to Make Automotive Radar Systems 'Hallucinate'

Engineers at Duke University have developed a system named "MadRadar," capable of manipulating automotive radar sensors, raising concerns about the security of radar technology in modern vehicles.

"In this work, we introduce MadRadar, a general black-box radar attack framework for automotive mmWave FMCW radars capable of estimating the victim radar's configuration in real-time and then executing an attack based on the estimates," the research team said.

The team was led by Miroslav Pajic, the Dickinson Family Associate Professor of Electrical and Computer Engineering, and Tingjun Chen, assistant professor of electrical and computer engineering.

All About MadRadar

MadRadar can deceive radar sensors into various scenarios, such as concealing the approach of an existing vehicle, generating a non-existent phantom car, or tricking the radar into believing a real car has suddenly altered its course.

It accomplishes these actions swiftly and without prior knowledge of the specific radar settings of the targeted vehicle, presenting a significant challenge to radar security.

The researchers emphasize that the existence of MadRadar highlights the urgent need for manufacturers to enhance the security measures of their radar systems to protect against potential misuse.

"Without knowing much about the targeted car's radar system, we can make a fake vehicle appear out of nowhere or make an actual vehicle disappear in real-world experiments," Pajic said in a statement.

"We're not building these systems to hurt anyone, we're demonstrating the existing problems with current radar systems to show that we need to fundamentally change how we design them," he added.

Hijacking Radar Sensors

Radar is vital in recognizing nearby moving vehicles for modern vehicles equipped with assistive and autonomous driving technologies. The research team noted that it supplements visual and laser-based systems to discern vehicles positioned in front of or behind the vehicle.

Owing to the varied array of vehicles on roadways employing radar, each with marginally distinct operating parameters, prior demonstrations of radar spoofing necessitated awareness of precise parameters for effective implementation.

Nevertheless, MadRadar transcends this constraint by promptly discerning a vehicle's radar parameters within a quarter of a second and subsequently initiating its own radar signals to mislead the targeted radar system.

"Think of it like trying to stop someone from listening to the radio. To block the signal or to hijack it with your own broadcast, you'd need to know what station they were listening to first," Pajic noted.

During the MadRadar presentation, the Duke team showcased the functionalities of their constructed radar-spoofing system, which can precisely identify a car's radar parameters in under a quarter of a second.

Once identified, the system can emit its own radar signals to deceive the radar of the target. The research is slated for publication in the 2024 Network and Distributed System Security Symposium, scheduled for February 26 to March 1 in San Diego, California, and is currently available on the arXiv preprint server.