Now would be the best time to cover up your webcam, disable your microphone access and put on a nifty tinfoil hat, but none of that will make any difference. Why? Because hackers can get to you through your phone’s vibration motor now. Well, sort of. A Solution is to talk in a high voice.
The “Vibraphone” research comes from Romit Roy Choudhury and Nirupam Roy, associate professor and Ph.D. candidate at the Electrical and Computer Engineering school of the University of Illinois at Urbana-Champaign.
And the idea itself is surprisingly very simple. A vibration motor is really like a very tiny speaker, and every speaker can be a microphone. Just think about it.
Okay, if you don’t understand how we got to this conclusion – A magnetic surface which has its position controlled by an applied voltage can also be arranged so that its position changes that voltage instead. That means one way it’s a speaker, the other way it’s a microphone.
So it’s not exactly fundamentally surprising that the vibration motor can be used to pick up sounds, the two wrote in their paper, and this PDF will be presented later this month at MobiSys in Singapore. The scary part? The fact that this is possible has been somewhat unexpected.
First of all, it should be made clear that this attack can’t be executed over the air or anything like that. The motor has to physically be rewired and apparently it’s a simple job that will only take a minute or two. There is a less invasive technique, though, it might be possible to exploit the power controller chip to collect the voltage fluctuation in the vibration motor, luckily this possibility hasn’t been investigated yet.
But, should this theory work, once the configuration is done the vibration motor will turn vibrations it picks up into voltage vibrations and that, can be reconstructed into waveforms. Presto! A microphone!
Unfortunately, turning a microphone into a speaker doesn’t make it a good speaker and vice versa. There is a limit to converting a vibration motor into a speaker; it’ll only be able to pick up vibrations on the low side, at about 2kHz. But comparing the original spectrograms to the recovered one’s will allow them to guess what other higher frequencies are present, which is still not great.
Take note; no one is recording an album here, and even in this very degraded state, the sound can still be recognized by people and voice recognition algorithms with decent accuracy. Some people got them to 80%, others to 60% and this is just the beginning, so someone will vastly improve the performance.
Luckily, for now, researchers confirmed that, if a sound doesn’t contain enough low frequencies the performance of the system will degrade. So the best defense we have against this new spy method is to talk in a high voice, all the time! Yes, your friends will laugh, people will think you’re crazy, but at least you won’t be spied on. Even though this will only kind of work if a soprano tends to be below 2kHz.
Then again, is someone going to sneak into your house, steal your phone, quietly disassemble it, rewire it and then listen to the garbled audio of you asking Siri for recipes? No, but that’s not the only way. Devices like fitness bands or smart watches, many of which doesn’t have microphones but do have a vibration motor, might perhaps be more enticing targets because no one thinks to carefully monitor what they say around their step tracker.
For now, researchers aren’t going into the whole spy direction but rather, more constructive applications. That deserves a round of applause. Looking forward to further research on this topic, or anything else that’s spy related.
Also take a look at these: Can we trust South African online shops?