For seniors living alone with compromised physical ability, unexpected events such as falls could prove to be fatal. Technological advancements such as cameras and other devices could help monitor these vulnerable individuals and provide them with the necessary help during such critical moments. Still yet, there exist a number of ‘blind spots’, which include areas like restrooms where cameras are seldom installed. Such blind spots can be found on virtually any site like at those where no light is available following accidental fires, making it extremely difficult for rescue workers to locate survivors.

The Mobile Networking Laboratory (MoNet Lab) headed by Professor Young-Joo Suh at the Department of Computer Science and Engineering, POSTECH, applies Artificial Intelligence to some of the core technologies that enable the 4th Industrial Revolution to eliminate such ‘blind spots’ in our daily lives. Researchers at the Lab leverage Internet of Things (IOT) and network technologies to detect elderly individuals at risk for falls and report such incidents to an emergency rescue center or hospital. Researchers are also working on technology that senses human activity to discern the intention behind their movements.

This is all made possible through the wireless signals found omnipresent in our modern society. If and when the trajectory of these signals is obstructed by human presence or by their movement, the signal experiences a slight distortion. With the help of AI, these distortions can discern human presence and movement. For instance, the hand movements involved in flipping through TV channels while washing dishes, alters the Wi Fi waves at home. As AI learns this pattern hundreds of times, it can also learn to recognize the gesture of a person waving.

Mobile signals provide numerous advantages over cameras. Cameras require light to function and a single camera is limited in the size of area it can recognize. They are also expensive and could potentially cause privacy issues. Mobile communication signals are more affordable than cameras and can even function in the dark. Not only can they penetrate though walls, they pose less of a privacy infringement risk, as their inability to clearly capture images ironically aids in avoiding privacy concerns.

Researchers at the Lab are also working to replace the Global Positioning System (GPS) which is useless in underground or indoor spaces with wireless signals that could potentially verify precise locations in indoor spaces. Magnetometers can be also used to measure the earth’s magnetic field. Since the earth’s magnetic field has distinctive values for every location, mapping the field through AI can help in navigating routes.

The MoNet Lab believes its greatest strength lies in its extensive research endeavors with businesses to produce outcomes readily applicable to daily lives. Recently, the Lab has developed AI technology to predict the lifespan of machines based on such information as vibration intensity or noise from sensors installed within smart factory machines as well as avatar anchor technology that is able to mimic human mouth movements. “We take great pride in generating more patents than papers as we value real-life applications”, Professor Suh commented, and added “We hope that our technology goes beyond mere application to contribute to saving lives and safeguarding public security”.