While the use of drones continues to grow, researchers from IOActive analyzed how to develop fault injection attacks against hardened Unmanned Aerial Vehicles (UAVs).
The experts focused on achieving code execution on a commercially available drone, supporting significant security features (i.e. the use of signed and encrypted firmware, Trusted Execution Environment (TEE), and Secure Boot), using non-invasive techniques. The research aimed at using electromagnetic (EM) side-channel attacks or EM fault injection (EMFI) to reach the goal. The researchers tested the attacks against one of the most popular quadcopter drones, DJI’s Mavic Pro.
In a first attack scenario tested by IOActive, the researchers attempted to retrieve the encryption key using EM emanations and decrypting the firmware. However, the tests revealed that the probability of a successful signature bypass was less than 0.5%.
A second approach tested by the researchers was the use of EMFI based on a previous study published by Riscure. Riscure demonstrated the use of a glitch to cause one instruction to transform into another and gain control of the PC register.
The researchers set up a test environment that included a laptop (used as a controller), a power supply, Riscure’s Spider (used to generate the trigger), an oscilloscope, an XYZ table, and the EMFI pulse-generator.
“After identifying a small enough area on the PCB, we modified the glitch’s shape and timing until we observed a successful result. The targeted process crashed” reads the report published by IOActive. “Our payload appeared in several registers. After examining the code at the target address, we determined that we had hit a winning combination of timing, position, and glitch shape.”
“Having successfully caused memory corruption, the next step would be to design a proper payload that achieves code execution. An attacker could use such an exploit to fully control one device, leak all sensitive content, enable ADB access, and potentially leak the encryption keys.” added the company.
IOActive shared their findings with the DJI team on 2023-04-04 which quickly offered their support and agreed on a publication date on 2023-05-04.
(SecurityAffairs – hacking, drones)