A new algorithm can identify changes in wireless signal transmission
Wi-Fi to monitor human respiration: Researchers at the National Institute of Standards and Technology (NIST) have developed a method to monitor people’s breathing based on small changes in Wi-Fi signals. The goal of the BreatheSmart deep learning algorithm is to detect if someone in the house has breathing problems. Wi-Fi signals are almost ubiquitous: they bounce and pass through surfaces as they try to connect devices to routers. But any movement will alter the signal path, including how the body moves as we breathe, which can change if there is a problem. Other researchers have explored the use of Wi-Fi signals to detect people and movement, but their approaches required dedicated sensing devices, with studies providing limited data. A few years ago, a company called Origin Wireless developed an algorithm that works with a Wi-Fi mesh network.
Similarly, NIST claims that BreatheSmart works with routers and devices already on the market. The scientists modified the firmware on a router to check ‘channel state information’ or CSI more frequently. CSI refers to signals sent to the router from a device, such as a phone or a laptop. CSI signals are consistent, and the router understands how they should be, but environmental deviations, such as signals influenced by surfaces or movement, change the signals. The researchers asked the router to request these CSI signals up to 10 times per second to understand better how the signal was being modified.
Wi-Fi for better living
The team simulated different breathing conditions with a dummy and monitored the changes in the CSI signals with a router and a standard receiving device. NIST researcher Susanna Mosleh developed the algorithm to make sense of the collected data. In a paper, the researchers noted that BreatheSmart correctly identified simulated breathing conditions 99.54 per cent of the time. Mosleh and Jason Coder, who head NIST’s research in shared spectrum metrology, hope developers can use their research to create software that can remotely monitor a person’s breathing with existing hardware. “All the ways we collect data are done on the software at the access point (in this case, the router), which could be done by an app on a phone,” Coder said. “This work seeks to illustrate how someone can develop and test their own algorithm. A framework to help them get relevant information.”
