Our world is full of signals, some of which we see and others which are invisible. These signals range from WiFi signals to brain waves. The pervasiveness of these signals results in networking the world through natural and man-made links. The aim of the Signals Kinetics group is to program these natural and man-made networks in order to extend human and computer abilities in communication, sensing, and actuation. We draw on tools from computer networks, signal processing, machine learning, and hardware design to uncover, analyze, and engineer these networks.
Research Projects
Cyber-Physical Security and Privacy (System & Network Security)
Fadel AdibIn the age of ubiquitous connectivity and the internet-of-things, our security and privacy have taken new dimensions. For example, how can we ensure that our locations are not being tracked from our cellphones? And, how can we prevent an unauthorized user from hacking into our smart home systems? Our research aims at developing primitives that can address these challenges. To do so, we explore intrinsically new security mechanisms that operate across all the computing stack to secure not only the bits but also the integrity of the sensed signals and to protect the privacy of the sensed environment.
Health Sensing (Human-Computer Interaction)
Fadel AdibToday’s health sensors (which monitor breathing, heartbeats, steps, etc.) require their users to wear them on their bodies. In contrast, our technologies can monitor human health without requiring the user to wear any device on his/her body. To do so, we capture and analyze wireless signals reflected off the human body; we then use these reflected signals to extract breathing and heartbeats without any physical contact with the human body. We are currently exploring techniques to remotely sense additional health metrics like blood pressure, oxygen saturation, and glucose levels. Monitoring these health metrics can render ICU (intensive care unit) vital sign monitors completely noninvasive and enable continuous monitoring of diabetes patients.
Programming Wireless Networks (Computer Networks)
Fadel Adib and Stacie SlotnickWireless networks — consisting of WiFi, LTE, RFIDs, and millimeter-wave devices — have become integral parts of our everyday lives. Our research explores how we can make these networks more robust, faster, and seamlessly mobile. It also explores how we can use these networks for purposes other than communication such as localization, sensing, and control.
Seeing Through Walls (Computer Vision)
Fadel AdibOur group develops technologies that can see through walls and perform motion capture through occlusions. To do so, we rely on wireless signals, like WiFi. These signals traverse walls and reflect off humans the wall before returning to a wireless receiver. We design and develop new algorithms and software-hardware systems that can extract these signals and analyze them to capture human motion from behind a wall.
Wireless Sensing for Drones & Agile Robots (Robotics)
Fadel AdibCan drones find objects which we’ve lost, even if those objects are hidden? Can agile industrial robots, like Baxter, use wireless sensing capabilities to identify objects inside closed boxes? Can personal robots, like Roombas, use their on-board WiFi to map indoor environments and feed this information to smart home automation tools? To answer these questions, our group explores placing wireless devices on robots to sense the wireless channel, extract RF reflections, and interpret them in the context of Computer Vision.
