New UCSB Program Can Count People Using Wi-Fi
As Signals Bounce, Objects Can Be Identified
A program developed at UCSB can count people using only the ubiquitous Wi-Fi signal and either the presence of bodies or their movement. Since the technique can count people even when they are carrying no Wi-Fi device, the program can be used passively in applications such as energy efficiency and search-and-rescue operations. The program was developed by Yasamin Mostofi, a professor of electrical and computer engineering at UCSB, and two UCSB graduate students.
In their testing phase, Mostofi and her students measured Wi-Fi signal strength as students wandered throughout the signal in an area about the size of a small apartment. They were able to count as many as nine people in an indoor and outdoor area. According to the researchers, measuring the Wi-Fi signal strength is key to their program. By placing two Wi-Fi cards at opposite ends of the testing area, Mostofi’s team was able to measure how much the Wi-Fi signal strength changed when people passed directly between the antennas, what the researchers called the “line of sight” method. The second technique used the fact that a person’s body can measurably scatter a Wi-Fi signal, even if they’re not directly in the Wi-Fi’s line of sight. Using a probabilistic math equation, the researchers used the signal strength measurements to calculate the number of people in an area.
The possible uses for the program are vast, they said. Since Wi-Fi inhabits nearly every home and business, the infrastructure to use the program already exists. First responders could use the technology to find people who have barricaded themselves behind locked doors. A green-building’s climate control could be hands free and based on a building’s occupancy. An open-minded person could come up with his or her own uses after downloading the free, online copy. “Stores can benefit from counting the number of shoppers for better business planning,” said Mostofi, in an interview with The Current, a UCSB publication. She plans to combine her work with previous research that analyzed stationary objects through walls using only a Wi-Fi signal.