1. We introduce MadRadar, a black-box attack framework for effective physical layer attacks on mmWave radars without prior knowledge of the victim radar's parameters (e.g., the chirp period and slope, and frame duration)

2.  We enable new black-box attack types by improving upon existing methods for estimating victim parameters

3. We demonstrate that mmWave radars are vulnerable to false-negative and translation attacks that effectively 'remove' or 'move' detections of existing objects in the victim's point cloud, respectively;

4. We demonstrate feasibility, and evaluate our attacks on multiple real-world case studies performed using a real-time implementation on the USRP B210 SDR platform.

David Hunt - Duke University

Kristen Angell - Duke University

Zhenzhou Qi - Duke University

Tiingjun Chen - Duke University

Miroslav Pajic - Duke University

Our paper can be accessed at this link: MadRadar paper .

Check out our MadRadar Git Repository!