Scientists prove WiFi signals can be used to track you from location to location

There are many ways to track a human, but researchers in Italy have devised a new method that involves creating a biometric marker for people based on how their body distorts WiFi signals.

2

The researchers are calling this new method of tracking "WhoFi," and detailed how it works in a paper that has yet to be published. The paper titled "WhoFi: Deep Person Re-Identification via WiFi Channel Signal Encoding" explains that re-identification is the primary goal of the new technology, which doesn't necessarily mean a person's identity, such as their name, but just that they are the same individual seen in one location and then in another.

The idea is that WhoFi could be used in video surveillance to maintain constant tracking of a subject without having to use any obvious identifiers such as items of clothing, features, etc.

How does it work? WiFi signals propagate through an environment and its signal is altered by the presence and physical characteristics of objects and people along its path. These changes are captured in the form of Channel State Information (CSI), which the researchers believe contain "rich biometric information".

The information being captured is the amplitude and phase of electromagnetic transmissions, which the researchers say are unique to every person and can be described accurately as "fingerprint" after the data is processed by a deep neural network and a unique data signature is created based on them.

According to the team behind WhoFi, the new tracking technique has an accuracy of up to 95.5%.

"The encouraging results achieved confirm the viability of WiFi signals as a robust and privacy-preserving biometric modality, and position this study as a meaningful step forward in the development of signal-based Re-ID systems," write the authors