Engineers open the door to next-generation wireless communications

In the early 2010s, LightSquared, a multibillion-dollar startup that promised to revolutionize mobile phone communications, declared bankruptcy after it couldn’t figure out how to prevent its signals from interfering with those of the GPS system.

Now, engineers at the University of Pennsylvania have developed a new tool that could prevent such problems from happening again: a tunable filter that can effectively block interference even at the higher frequencies of the electromagnetic spectrum.

“The hope is that this will enable the next generation of wireless communications,” says Troy Olson, an associate professor of electrical systems engineering (ESE) at the Pennsylvania Institute of Technology and senior author of the November 2010 paper. Nature Communications Let me explain about filters.

The electromagnetic spectrum itself is one of the most precious resources in the modern world. Only a small portion of the spectrum, primarily radio waves, making up less than one billionth of a percent of the total spectrum, is suitable for wireless communication.

Bands in that portion of the spectrum are tightly controlled by the Federal Communications Commission (FCC), which only recently made available for commercial use the Frequency Range 3 (FR3) band, which includes frequencies from about 7 GHz to 24 GHz. (One Hertz is equal to one oscillation of an electromagnetic wave passing through a point in one second, and one Gigahertz (GHz) represents one billion such oscillations per second.)

Until now, wireless communications have primarily used lower frequency bands. “Right now we’re operating from 600MHz up to 6GHz,” Olson says. “That’s 5G, 4G, 3G.”

Wireless devices use a different filter for each frequency, so to cover all frequencies or bands, a large number of filters taking up a significant amount of space are required. (A typical smartphone has over 100 filters to keep signals from different bands from interfering with each other.)

“The FR3 bands will most likely be deployed in 6G or Next G,” Olson says of the next generation of cellular networks. “Currently, small filter and low-loss switch technology has very limited performance in these bands. Having filters that are tunable across these bands would eliminate the need to have another 100+ filters in a phone with different switches. Filters like the one we’ve created are the most practical way to use the FR3 bands.”

One problem with using higher frequency bands is that many of the frequencies are already reserved for satellites. “Elon Musk’s Starlink operates in these bands,” Olson points out. “The military is already locked out of a lot of the lower frequency bands. They’re not going to give up their radar frequencies or satellite communications, which are in these bands.”

As a result, the Olson lab, in collaboration with colleagues Mark Allen, Alfred Fittler Moore Professor at ESE, Firouz Aflatouni, Associate Professor at ESE, and their groups, designed a tunable filter that allows engineers to selectively filter out different frequencies using this filter, eliminating the need for separate filters.

“Tunability is going to be really important,” Olson continues, “because at these higher frequencies, there aren’t always blocks of spectrum that are dedicated just for commercial use.”

What makes the filters tunable is a unique material called “yttrium iron garnet” (YIG), a mixture of yttrium, rare-earth metals, iron, and oxygen. “What’s special about YIG is that it propagates magnetic spin waves,” Olson says, referring to the type of waves that are generated in magnetic materials when electrons spin in sync.

The magnetic spin waves generated by YIG change frequency when exposed to a magnetic field. “By adjusting the magnetic field, the YIG filter achieves continuous frequency tuning over an extremely wide frequency range,” says Xinyu Du, a doctoral student in Olson’s lab and first author on the paper.

As a result, the new filter can be tuned to any frequency between 3.4 GHz and 11.1 GHz, covering much of the new territory the FCC has pioneered in the FR3 band. “We hope to demonstrate that a single, highly adaptable filter is sufficient for all frequency bands,” Du says.

In addition to being adjustable, the new filters are very small, about the size of a quarter coin, as opposed to the previous generation YIG filters, which resembled a large stack of index cards.

One of the reasons the new filter is so small, and potentially even found in future mobile phones, is that it consumes very little power: “We pioneered the design of zero-static-force magnetic bias circuits,” Du says, referring to a type of circuit that generates a magnetic field without requiring any energy other than occasional pulses to realign the field.

YIG was discovered in the 1950s, and YIG filters have existed for decades, but the combination of this new circuitry and ultra-thin YIG films microfabricated at the Singh Nanotechnology Center dramatically reduced the power consumption and size of the new filter. “Our filter is one-tenth the size of current commercially available YIG filters,” Du says.

Olson and Du plan to present the new filter at the Institute of Electrical and Electronics Engineers (IEEE) Microwave Theory and Technology Society (MTT-S) 2024 International Microwave Symposium in Washington, DC in June.