The TempestSDR system captures electromagnetic signals leaking from HDMI cables to reconstruct video content on a remote display. While the initial signal quality is low, researchers have introduced an AI-powered tool called DeepTempest to enhance the output.
Developed by Federico Larroca from the University of the Republic in Montevideo, Uruguay, DeepTempest improves text recognition by reducing errors by more than 60 percent compared to previous methods. When recovering text, the system currently maintains an error rate of approximately 30 percent.
The technique relies on capturing radio signals via an antenna and a HackRF radio. Success depends on factors such as cable shielding, environmental interference, and the proximity of the attacker to the target device.
The term Tempest refers to a US National Security Agency codename for monitoring electromagnetic leaks and implementing countermeasures. Experts recommend using high-quality shielded cables to mitigate the risk of such data interception.
