NSF backs project to create next-gen wireless devices
ITHACA, N.Y. – Two Cornell University researchers are looking into a new way to meet the growing demand for wireless services in the U.S.
Amal El-Ghazaly and Alyosha Molnar, from the School of Electrical and Computer Engineering (ECE) at Cornell, have received an $880,000 grant from the National Science Foundation (NSF) to design a new class of radio devices capable of operating across a large portion of the growing wireless spectrum, while adaptively suppressing interferences.
“The electromagnetic spectrum has a lot of different frequencies that are accessible for communication,” said El-Ghazaly, an assistant professor of ECE. “But there are so many more users than there are frequencies.”
To meet growing demand for wireless services in the U.S., the Federal Communications Commission has periodically made an ever-widening range of frequencies in the electromagnetic spectrum available for unlicensed usage. This creates opportunities for innovation in the way devices like cellphones and computers use the spectrum for various types of communications.
As the usable spectrum becomes more congested, wireless devices may be forced to share frequency bands or use frequencies tightly packed together, leading to interference. New wireless systems must become more robust against interference to take advantage of the increased bandwidth.
To spur innovation among researchers, the NSF created the Spectrum and Wireless Innovation enabled by Future Technologies (SWIFT) program. SWIFT is looking for innovations in transmitters, receivers and spectrum coexistence, which refers to two or more applications using the same frequency band at the same time without adversely affecting one another. Spectrum coexistence has received less attention from researchers.
El-Ghazaly and Molnar, along with Bernd-Peter Paris, associate professor of electrical and computer engineering at George Mason University, have a unique approach to coexistence in their SWIFT-funded project: Use the entire range of frequencies and apply adaptive, tunable filtering to reject any interference. Their project is titled “Adaptive Interference Rejection with Synthetic Channel Diversity.”
The team will develop a radio receiver architecture capable of operating across a large portion of the wireless spectrum while suppressing interferences as they arise. A new algorithm will allow the receiver to adaptively adjust its interference response as needed using digital signal processing.
“Your cell phone has certain bands it can use and some it can’t,” said Molnar, an associate professor of ECE. “For any given band you want to use, you need a filter that’s engineered to just let that band through and block everything else.”
But those filters are fixed, or only very weakly changeable, he said, when you build the device.
“You can’t really use the spectrum efficiently because you have to commit in advance [to] which bands you want to use,” said Molnar. “The approach we’re taking is, rather than building these very narrow filters that only pass certain things, we’re going to pass everything through.”
But when a device receives all the frequencies, it’s also taking in a lot of interference, overwhelming the receiver. The key: designing a device to enhance the signal in the frequency you want to receive and suppress interference at others. Such a device could take in a much wider band but also be able to filter out interference to focus on the desired signal.
“Any sort of band-limiting filter needs to have at least a couple of different components,” El-Ghazaly said. “When the components are tunable, we’re able to slightly change the values of the inductor and the capacitor so that we can adjust the network as needed to be optimal for each new set of signals and interferences coming in.”
The algorithm analyzes the signals and calculates how to tune each of the values in real time, to get the best possible signal.
This type of device would remain usable even as the spectral environment changes. As regulators make additional bands available, such devices would already be capable of utilizing them. Likewise, bands that are in use only in certain locations, or at certain times of day, would become available.
“If you are near a satellite station during certain times of day when the satellite is overhead, you can’t use that band,” Molnar said. “That only happens a few times a day in certain locations, but the rest of the country is blocked from that band.”
With his team’s new adaptive architecture, those frequencies could become usable. The new receiver will be dynamic and tunable, the researchers said, responding in real time to whatever bandwidth is available and unconstrained by its own hardware.
The project includes plans to educate and train rising engineers, at both the graduate and undergraduate levels, to think holistically about the components and operation of wireless systems and design robust receivers for the future.
Some of the initial concepts foundational to this project were funded by the Semiconductor Research Corporation, through the Joint University Microelectronics Program.