Having previously used Wi-Fi signals to look through walls, a team of researchers in professor Yasamin Mostofi's lab at the University of California, Santa Barbara, has now turned the wireless signals to the task of counting the number of people walking in a particular area – even if they aren't carrying any Wi-Fi-enabled devices.
The technology is basically the same as that previously used by Mostofi's team to look through walls. But rather than identifying the position and outline of stationary objects within a walled structure, the changes in the strength of the sent and received Wi-Fi signals was used to estimate how many people were walking in a given area.
To put the technology to the test, the researchers used two Wi-Fi cards, placing one at each end of a target area of roughly 70 sq m (753 sq ft). Using the received power measurements of the link between the two cards, it was possible to estimate the number of people walking within the area using a probabilistic mathematical framework.
This largely relied on two key things – the drop in the signal as a person crosses the direct line of sight between the two cards, and the scattering of the signal being reflected back to the receiving card when the person is not in the direct line of sight, which is known as multi-path fading.
The researchers have successfully tested the technology with nine people in both indoor and outdoor settings, with Mostofi believing it could find applications in energy efficiency and search and rescue given the near-ubiquity of Wi-Fi signals in many areas.
In terms of energy efficiency, she suggests the technology could be used to estimate the number of people in a building, and adjust the heating and cooling automatically. Similarly, this kind of information could help with search and rescue operations.
Mostofi says she plans to eventually bring the two Wi-Fi-based technologies together, which could result in a system that could estimate the number of people moving in a given area, but also locate people behind solid walls.
The video below explains the head-counting Wi-Fi technology, with the team's paper scheduled for publication in the Institute of Electrical and Electronics Engineers Journal on Selected Areas in Communications.
Source: UC Santa Barbara