Platypus system from Microsoft Research can identify humans from their electrical fields

Arif Bacchus

From a smart tattoo to swim training technology, and even looking at DNA to store data, Microsoft Research is always busy coming up with new technological advancements. Today, we’ve come across yet another Microsoft Research project (via Neowin), and this time, it involves identifying humans from their electrical fields.

In a full 14 page document, Microsoft Research describes the Platypus system as the first system to, “localize and identify people by remotely and passively sensing changes in their body electric potential which occur naturally during walking.” The document also describes that the system involves using uses three or more electric potential sensors (seen below) with a maximum range of 2 m. This means that Platypus is a passive tag free system that does not require any special hardware such as a smartphone, smartwatch, or other wearable devices.

Screen Shot 2016-06-02 at 9.52.50 AM
A look at the sensors in Platypus

The Microsoft Research team managed to use Platypus to locate a human in the room within an accuracy of 0.5 feet. Based on short walking sequences of 5 s, they were also able to identify four users of Platypus within the accuracy of 94 %, and 30 users with an accuracy of 75 %.

Despite these results, at the end of the document detailing the system, Microsoft Research notes that Platypus has some limitations. These include the need for person-specific sensor calibration that is independent of the user’s height/weight/clothing, as well as the presence or absence of grounded electrical objects in the environment.