Identifying People Via WiFi Waves!

FreeSense uses the radio-wave frequencies of WiFi to identify individuals.

The world is undoubtedly contracting every passing month — as everybody is connected to each other in the world of Facebook, Connected Devices and via the Internet of Things.

Smartphones, smart TVs, internet enabled surveillance cameras, cute and interactive kids’ toys through which children can send electronic messages to their parents, are all smart appliances that, just with an application installed on your smartphone can be controlled directly through the smartphone.

This obviously has numerous advantages such as being convenient, not requiring the addition of any extra cameras, microphones, scanners, sensors, cards, wearable tags to an ecosystem of devices which is over-populated. Basically all things good and hassle free.

Technology being the double edged sword that it is though, all these connected objects are significantly risky as they are prone to hacking and can pose a real threat to one’s security and privacy.

Instances from the real world to back this premise are numerous, security cameras used to spy on users being one of them.

The only way out of all this is to improve the modes of authentication to establish a secure, smart network in the future. While there are many methods like facial recognition, gait recognition and using biometric data like fingerprints that are prevalent or are being proposed; they are still bound by limitations, in the sense that these methods require additional scanning devices like scanners which in turn, are again prone to hacking.

However, a team of researchers from China’s Northwestern Polytechnic University have proposed a seemingly better alternative to Internet of Things authentication.

If reports by Motherboard are to be believed then the method dubbed as FreeSense uses near-ubiquitous, radio-wave frequencies (RF) of WiFi to identify individuals. This is done by locating the unique perturbation pattern produced when the individuals move around and intrude these signals, as per their unique body shape and gait.

You must be wondering as to how does this system track unique perturbation patterns.

This is done by tracking changes to a WiFi signal’s channel state information (CSI) as it disseminates through the space between a transmitter and a receiver.

Due to the difference of body shapes and motion patterns, each person can have specific influence patterns on surrounding WiFi signals while she moves indoors, generating a unique pattern on the CSI time series of the WiFi device”, explained the researchers in their paper. “FreeSense…is nonintrusive and privacy-preserving compared with existing methods [of human identification]”.

The team of researchers also elaborated on the mechanism of FreeSense: “Specifically, a combination of Principal Component Analysis (PCA), Discrete Wavelet Transform (DWT) and Dynamic Time Warping (DTW) techniques is used for CSI waveform-based human identification”.

The researchers conducted an experiment whereby nine individuals were tracked in a furnished, 322-square-foot smart home environment, making use of a conventional WiFi router and laptop. The researchers initially prepped their system with data garnered from each each individual’s gait as they walked across the room in a straight line. Next, the researchers repeatedly made the subjects participating in the test to walk across the room in order to figure out the accuracy of the system in identifying each individual.

The results?

The system performed unexpectedly well. With just one person in the room, it was capable of differentiating the individual 75% of the time; in case of two people the accuracy level went up to 95% of the time. when the number of people inside the room was increased to six, FreeSense was able to successfully identify each person with the accuracy rate of 90% — which is unbelievable, one must say — keeping in mind that some individuals may have been harder to differentiate with others if they happened to have a similar gait and/or body shape.

The researchers are now putting their efforts in the direction of increasing the performance of the system in various scenarios.

Our Take

Identification through Wi-Fi seems like a good solution, although we can expect other better proposals to come along in the next few years.

Use of Wi-Fi signals for authentication seems to have some benefits like it is a convenient, device-free alternative which doesn’t require any extra devices like cameras, scanners, microphones, wearable tags, sensors or cards, instead utilising the already existing WiFi infrastructure to function.

Also these persistent WiFi signals don’t require adequate light and line of sight to aid the identification of users. It is definitely less intrusive and tiresome as compared to other methods like facial recognition and fingerprinting and looks suitable for use in domestic scenario, or for small-scale uses like in homes, or in assisted living environments equipped with smart appliances and technologies.


Originally published at Chip-Monks.