Radiofrequency machine learning

Our researchers are harnessing AI to develop wireless devices that can learn to share the RF spectrum optimally, in real time—and distinguish between legitimate IoT devices and spoofers.


Paper: Exposing the fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting

Data Set Download


While blockchain technologies ensure that digital transactions remain transparent and verifiable to all parties, these systems require vast amounts of energy. We’re harnessing AI to enhance their efficiency for use in IoT systems.

Intrabody medical systems

For patients with chronic diseases, we are devising tiny sensors and actuators that transmit power and data using unconventional wireless techniques that are safer, more secure, and more energy efficient.

Unmanned aerial and underwater IoT networking

Our goal is to enable drones and other technologies to synchronize their sensing, communications, and movement autonomously, and to learn to foil adversaries by listening to the wireless spectrum and continuously changing their modes of communication.

Optimizing data storage networks

Caching networks curb server traffic for data providers and speed data delivery for users, yet we have no objective, mathematical standards for measuring their efficiencies. We aim to optimize cache network performance given a set of design objectives (such as optimizing data throughput or minimizing costs) and assess our algorithms under real-life scenarios.

Terahertz-band communication for 6G networks

Harnessing the enormous bandwidth available at minute THz-band frequencies will be a key advance in wireless technology in the next decade, but many roadblocks must first be overcome. Our Ultra-broadband Nanonetworking (UN) Lab is making fundamental contributions to the science and technology of wave generation and reception at the nanoscale.