Researchers find way to listen in on headphones from afar
Researchers unveiled InjectEave, an electromagnetic injection side-channel attack that recovers audio from headphones and landline phones up to 30 meters away through walls.
Researchers from HKUST and The Hong Kong Polytechnic University presented InjectEave at USENIX Security 2026, an active EM side-channel attack that injects 0-9 MHz RF signals into nonlinear components such as amplifiers, ADCs, and power converters to modulate and leak target audio. Tested on 11 off-the-shelf devices including Sony, HP, Philips, Apple, and Xiaomi products, the attack recovered intelligible headphone audio from up to 30 meters with an RF amplifier, generally 1-6 meters otherwise, including through walls. The attack uses commodity gear such as a USRP B210 SDR and a Siglent spectrum analyzer, is immune to digital defenses like encryption, and is only partially mitigated by shielding, twisted-pair wiring, and filtering.
- InjectEave injects 0-9 MHz RF signals into nonlinear hardware to induce detectable audio leakage
- Recovered intelligible headphone audio from up to 30 meters away, including through walls
- Verified on 11 commercial devices from Sony, HP, Philips, Apple, Xiaomi, and others
- Requires commodity equipment: USRP B210 SDR, antennas, and Siglent SSA3075X Plus spectrum analyzer
- Immune to encryption and masking; shielding and filtering only reduce exposure
- Products
- InjectEave
- Organizations
- HKUSTThe Hong Kong Polytechnic University
- Countries
- China
From the original article
3 min read at theregister.comResearchers based in China have devised a way to eavesdrop on signals handled by analog components in devices such as headphones, landline handsets, and smart devices by injecting electromagnetic (EM) signals. The technique, referred to as InjectEave, is not simply listening in on a low-frequency analog signal.
By Thomas Claburn · theregister.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.