Connect with Joseph Steinberg

Cool Tech

Even Without Seeing You, People Can Read Your Lips, Identify You, and Read Your Writing – Thanks To Wi-Fi

Even Without Seeing You, People Can Read Your Lips, Identify You, and Read Your Writing – Thanks To Wi-Fi

Did you know that by analyzing Wi-Fi signals governments and others can spy on you physically, seeing when and how you move, and monitoring what you say or write?


In August of 2016, a group of researchers at Northwestern Polytechnical University in China shared information online about a system that they claim can, with sometimes greater than 90% accuracy, identify people based on how their bodies affect Wi-Fi signals. Their work followed a similar study by Australian and British researchers, other research done in Asia, and work done at MIT that found that by analyzing Wi-Fi signals one can detect if someone in a room nearby is waving his or her hand.

How is any of this possible?

Wi-Fi technology, of course, uses radio frequency transmissions. Anytime that you are in an environment in which there is Wi-Fi, it is not just electronic devices that the Wi-Fi affects; as Wi-Fi’s radio waves hit your body you absorb some of them, and cause others to bounce off of you in various directions. By analyzing how a Wi-Fi signal changes when a particular human moves through it, therefore, researchers can determine that individual’s position, shape, and movement. Armed with sufficient knowledge in advance – such as what a particular person looks like and/or how he or she walks – someone with an analysis system can often garner sufficient data from Wi-Fi signals in order to positively identify a person. The Chinese researchers who recently studied the matter, for example, claim that, in their tests, in which they pre-loaded information about various people’s body shapes, their monitoring system could identity which of two people walked across its line of sight with 95% accuracy. When given six possible choices, the system was right 89% of the time.

But the privacy risks are even greater. Careful analysis of the impact on Wi-Fi signals of a human moving within them can also allow outsiders to determine what a person is saying by reading that person’s lips, or to read what a person is writing with a pen or typing with a keyboard (after all, all such activities impact Wi-Fi signals).

The Wi-Fi access point or router does not even need to be in the same room as the person being monitored. As long as there is sufficient signal to analyze the impact of motion, all sorts of spying is possible. To make matters worse, the person being monitored might not even know that there is Wi-Fi present in the area; the person doing the monitoring could be the one supplying the Wi-Fi signal.

Of course, Wi-Fi-signal based monitoring could also be used to improve the world – for example, by detecting home invasions in the middle of the night, or by alerting emergency services if an elderly person falls and cannot get up. Wi-Fi analysis could ultimately be used to trigger alarms in cases of other forms of medical emergencies; the aforementioned MIT researchers claim that by analyzing changes in Wi-Fi signals one can, with almost perfect accuracy, remotely monitor the breathing and heart rates of humans present in any area in which sufficient Wi-Fi signal is present.

As far as we know, all of the aforementioned Wi-Fi scanning and analysis is still in the research phase. Governments, of course, might have already secretly implemented such technologies. In any case, keep in mind that wherever there is Wi-Fi or any other radio signals, it is possible that someone could exploit them to “see” things that most folks assume are invisible. Top secret conversations, for that reason, should take place in environments shielded from radio transmission leaks. In fact, for such reasons, and to prevent possible leaks of data via similar analysis of electromagnetic fields, the government implements secure, shielded areas in its secure facilities.

(An earlier version of this article originally appeared in August of 2016 in Inc.)

Continue Reading

More in Cool Tech




* indicates required