ISSN 2330-717X

Pursuing Next Big Leap In Radar And GPS Systems


Xu Yi, assistant professor of electrical and computer engineering at the University of Virginia, is leading a research team pursuing the next big technological leap in radar and global positioning systems – part of a national effort to advance these technologies using photonics.


Yi’s ambition is to translate the purity and stability of high-frequency optical signals into the microwave regime where defense capabilities for positioning, navigation and timing typically operate. To put this in perspective, the system they plan to develop will function up to 110 gigahertz, frequencies 20 times higher than WiFi and three times higher than 5G.

Yi formed a team with colleagues in the UVA School of Engineering and Applied Science’s Charles L. Brown Department of Electrical and Computer Engineering, the University of California Santa Barbara, Morton Photonics and Honeywell to achieve this goal. They earned a $2.4 million three-year grant from the Defense Advanced Research Projects Agency’s GRYPHON program, which stands for Generating RF with Photonic Oscillators for Low Noise.

The project builds on Yi’s achievements in photonics, one of the department’s research strengths. Yi and teammates Andreas Beling, professor of electrical and computer engineering, and Steven M. Bowers, associate professor of electrical and computer engineering, have achieved breakthroughs in high-speed light detection and ultra-low noise circuits to bring photonic technology closer to reality.

“Conventional approaches start with low frequencies and multiply up, to arrive at higher frequencies that are useful for sensing and communications,” Beling said. “Our project tackles the problem from the other side, to start high and divide down, converting light into radio waves.”

Radars, the Global Positioning System and space missions rely on microelectronic systems. Yi’s team is focused on the component at the heart of these systems, the microwave oscillator. The oscillator produces a high-frequency electromagnetic wave or energy pulse to coordinate and schedule data flows through high-speed digital systems, synchronize linked systems and convert signals from high to low frequencies.


Yi specializes in a specific type of photonic device called a microresonator-based frequency comb, or microcomb. The microcomb efficiently converts photons from single to multiple wavelengths. Yi’s innovations in optical frequency division offer a pathway to make a chip-size, low-noise system that is continuously tunable over a very large range.

An ideal oscillator provides a perfect signal at a single frequency. Because real-world systems such as military radars and commercial 5G systems operate at variable frequencies, they are much less stable, a limitation generally referred to as phase noise. 

“Phase noise — how much the signal wiggles — is the metric everyone cares about,” Beling said. “Once you have a super-stable signal that is generated in a small, integrated package, this opens up new possibilities in applications like communications, positioning and ranging.”

“In this specific setting, if we’re chasing low noise, photonics has a key advantage over electronics,” Yi said. “With this program funding, we have the opportunity to form a very good team and push that to the limit, to see how far this technology can go.”

To work with deployed systems, optical signals must be converted into the electrical domain, similar to a solar cell that converts light to current. Beling’s group has demonstrated this capability in stand-alone devices. His photodetectors offer proof-of-concepts for output power or signal strength; linearity, which is another way of expressing a clean signal; and bandwidth.

“We have this established photodetector technology at UVA,” Beling said. “But instead of daisy-chaining discrete components, we envision a signal generator or synthesizer fabricated as one integrated device.”

Bowers will take point on systems integration, which involves both photonic and electronic devices and their interaction. Bowers’ primary task is to develop an opto-electronic feedback system to achieve continuous tunability with improved phase noise performance.

“A tuning fork is a really resonant device,” Bowers said. “When you strike a tuning fork, it produces one tone. If you have a tuning fork for the ‘a’ note, and want to also tune the ‘b’ note, you need a whole new fork. The photonics teammates can create these really phenomenal high-quality-factor tuning forks, these highly resonant devices, but we need to monitor their output and return a feedback signal to make sure the signal remains stable.”

Bowers leads the Integrated Electromagnetics, Circuits and Systems research group, whose members will design and integrate opto-electric control circuitry and control algorithms to meet this requirement. Bowers is also an affiliated faculty member of the Link Lab, UVA’s world-class center of excellence in cyber-physical systems.

“What’s really hard about this program is getting that clean of a signal and being able to tune it across a wide range of frequencies,” Bowers said. The team’s mid-point goal is to generate a signal from one to 110 gigahertz with a one-gigahertz resolution. Bowers’ feedback and control circuitry will extend the resolution to the right of the decimal point, to one-hertz resolution or more.

Bowers’ second task is to develop electronic frequency dividers to extend the benefits of phase noise performance to the one-gigahertz level—to optically produce a clean 32-gigahertz signal and divide it by 32 to get a one-gigahertz signal that is equally stable, meaning no additional noise.

In sum, Yi and other teams that have earned GRYPHON grants are asked to offer at least an order of magnitude leap in one of three target metrics: size, phase noise and frequency span. This combination of features is unprecedented today and will establish new source technology that is expected to transform the types and capabilities of military and commercial radar and communication systems.

Leave a Reply

Your email address will not be published.