At the Edge of Wireless Innovation

In 1996, researchers at AT&T made a breakthrough in wireless technology.

They figured out how to improve the speed, efficacy and reliability of wireless communication by correlating signals from multiple antennas, as opposed to just one. This “space-time coding” technology remains integral to mobile phone calls to this day.

The researchers’ landmark paper in 1998 on their work has since been cited more than 10,000 times. Robert Calderbank, who led AT&T research, and Vahid Tarokh, who led the wireless and signal processing department at AT&T, later joined Duke University in the 2010s as faculty members in the Department of Electrical & Computer Engineering (ECE).

Three decades after that important innovation in wireless technology helped usher in 2G and 3G technologies, the industry is now rapidly moving toward 6G.

Duke Engineering is working to play a key role in developing the next generation of wireless technologies thanks to a growing roster of experts—including many beyond Calderbank and Tarokh—and collaborations across campus.

“Duke has a unique team of theorists and experimentalists who can talk in each other’s language and closely collaborate,” said Hai “Helen” Li, the Marie Foote Reel E’46 Professor and chair of ECE.

Experimental Collaborations across Duke

For wireless signals to work, they must navigate the idiosyncrasies of the real world, like topographical features, dense urban spaces or quickly moving objects. Tarokh said the theoretical limits and mathematics in telecommunications are largely well understood, but the greatest challenges remain in implementation.

“The question becomes how to carefully engineer network infrastructure given the numerous real world challenges involving geography, spectrum sharing, legal issues and more,” Tarokh said.

Over the years, Duke has not only built up its theoretical prowess in telecommunications, but also its experimental expertise. Tingjun Chen, assistant professor of ECE, bridges both areas with focuses on the integration of wireless and fiber optic networks, the optimization of signal transmission and the development of digital twins for signal propagation. He is also working on incorporating sensing capabilities into this technology.

“At Duke, we not only have in-lab testbeds but also a campus-scale one that includes commercial-grade private LTE/5G networks and a county-scale fiber network,” Chen said. “This allows us to test ideas out—such as new signal designs, new algorithms and new machine learning methods for communication and sensing—at scale. I don’t know of any other university in the country with those capabilities.”

Chen commonly collaborates with Calderbank by testing his lab’s customized wireless signals on network resources provided by Duke’s Office of Information Technology (OIT), which maintains hundreds of miles of fiber optic cables connecting Duke’s East and West campuses and the Chesterfield Building in downtown Durham.

“OIT is committed to making our network an asset to researchers,” said John Board, associate professor of ECE and associate chief information officer at OIT. “Our infrastructure is in service to the research mission, while also serving its operational functions.”

The collegiality, collaboration and interdisciplinarity at Duke allows faculty members to explore telecommunications from a variety of viewpoints. Galen Reeves, associate professor of ECE and statistical science, frequently works with Henry Pfister, the Jeffrey N. Vinik Professor of ECE, on information theory, which provides a mathematical framework for studying how information is processed, stored and communicated.

“Bringing together expertise from related areas in machine learning and statistics leads to new insights and advances in telecommunications, ” Reeves said.