May 2, 2023
Whether it’s a Fitbit, an Apple Watch, or an Oura Ring, wearable computers have gone mainstream and are used daily by many people around the world. At the University of Pittsburgh School of Computing and Information, however, an assistant professor in the Department of Computer Science, Dr. Longfei Shangguan, is researching “earable” computers. Shangguan described earable computers as “embedded sensing devices that can be put on or near an ear to sense physiological activity, such as heart rate or breath rate.” Much like other wearable computers, these sensor devices are worn on the body to collect data about the person wearing them, often taking the form of an earbud, among other styles of earphone.
Shangguan noted that while other researchers are focused on a “sensor-adding paradigm” for developing this technology, he opts for a different approach. “We are reducing sensors,” he said. “We want to transform all existing non-smart earphones into a versatile sensor, in order to sense physiological activity. For example, we want to transform a speaker on our commodity earphone into a sensor to detect things like heart rate.” To achieve this, Shangguan said that he “uses the earphone itself as the sensor. We build the hardware and then we transform all existing earphones into smart ones, and then we can use them for earable computing.” Thus, rather than having to create a number of components to assemble an earable computing unit, Shangguan and his team can retrofit already manufactured earphones with sensors to achieve the same functionality.
In Shangguan’s estimation, there is a real advantage to refurbishing existing earphones to function as earable technology. “If you consider underdeveloped countries, the ‘sensor-adding paradigm’ may be inaccessible for many. With our approach, it is much cheaper.” He added, “Recently we got funding from the Pitt Momentum Agency. We got some money to support our purchase of materials to produce earable computing devices in the underdeveloped world.” Shangguan said that the overall cost to convert existing earphones into “an earable computing platform that supports physiological sensing” would be “less than five dollars” per unit.
Shangguan said that his team is currently building an earable computing platform that will allow novices to quickly develop prototypes of their research ideas without touching the underlying sensors, circuits, and drivers. “We will make this platform open source and hopefully it will lower the barrier of entry to the earable computing research community.”
Looking forward, Shangguan said that the future of earable computers and other devices that fall under the “Internet of Things” category is “networking almost everything in our ambient environment together.” He reminded us, however, that, “We are far away from this mission.” In the meantime, Shangguan said he will continue researching “three fundamental problems” connected to earable computing and similar technologies: “sensing, communication, and computing.” Achieving improved “sensing accuracy, computing efficiency and power efficiency” are three key steps along the way, Shangguan said.
--Daniel Beresheim