DoD gives $2.7M to fund a 5G spectrum-sharing initiative in Salt Lake City

Zylinium Research will develop an AI-based spectrum sharing system alongside the Department of Defense (DoD) and the Platforms for Advanced Wireless Research (PAWR) Project in Salt Lake City.

Sue Marek, Special Contributor

May 28, 2020

3 Min Read
Light Reading logo in a gray background | Light Reading

Thanks to $2.7 million in funding from the Department of Defense (DoD), the Platforms for Advanced Wireless Research (PAWR) Project will soon be conducting tests in Salt Lake City to see if two mobile operators can occupy the same spectrum in the same channel without interference.

The PAWR Project, which consists of 35 industry partners including US Ignite and Northeastern University, will conduct tests on a live 5G New Radio network using Citizens Broadband Radio Service spectrum in the 3.5GHz band.

The test will be in Salt Lake City and it will use a customized emulator from Zylinium Research. The emulator, called the Zylinium Spectrum Exchange, uses artificial intelligence to coordinate spectrum usage at a highly granular level.

Zylinium Research was one of the winners of DARPA's Spectrum Collaboration Challenge – a three-year open competition in which teams from around the world competed to solve spectrum management problems using artificial intelligence (AI). The competition ended last October and DARPA handed out awards at last year's MWC trade show in Los Angeles.

According to Thyagarajan Nandogopal, deputy division director for the Division of Computing and Communication Foundations at the National Science Foundation, DARPA launched the spectrum collaboration challenge as a way to break assumptions on how spectrum is allocated.

DARPA even built Colosseum, a radio-frequency emulation testbed at the John Hopkins University Applied Physics Laboratory, just so the Spectrum Collaboration Challenge teams could test their emulators and calculate how the signals would behave in certain environments. "This showed promise for this free-form allocation of spectrum resources, but DARPA wanted to know how it works in the real world," Nandogopal said while speaking at the 5GX Connect Virtual Summit Wednesday. "That's why we now have a testbed in Salt Lake City that is ready to go."

Zylinium Research is partnering with the PAWR and the University of Utah to transition the company's spectrum optimization technology from an emulation environment to the city-scale testbed.

The funding from the DoD will make it possible for Zylinium to create a full 5G NR software profile that it can then use with its Spectrum Exchange to determine how finite amounts of spectrum can be allocated using artificial intelligence.

The Zylinium Spectrum Exchange, however, should not be considered a replacement to the existing Spectrum Access System that is currently used to make sure different types of users in the CBRS spectrum band – incumbents, Priority Access Licenses users and General Authorized Access users – don't interfere with each other. Instead, this Spectrum Exchange is intended to allow more spectrum sharing within these spectrum tiers and create greater spectral efficiency.

Nandogopal said that while this is the first test, the National Science Foundation and DARPA are planning other tests as well. There will be one test focused on millimeter wave (mmWave) spectrum in New York City and also a test that focuses on autonomous cars and drones. The groups also want to create a fourth test that will examine rural broadband delivery.

However, Nandogopal said that this first test in the CBRS spectrum band is particularly critical because it involves mid-band spectrum, which is a very desirable spectrum range. And he added that individual companies can try to test this type of spectrum sharing technology using AI on their own but it's important to do it in a testbed environment to get unbiased data.

Nandogopal also said that, should this test in Salt Lake City be successful, it could have significant implications to companies in the telecom industry. "Telecom networks have largely been governed by manual processes relying on human knowledge," he said. "Early studies show that this offers significant gains. AI can be a vital tool for wireless networks."

— Sue Marek, special to Light Reading. Follow her @suemarek.

About the Author

Sue Marek

Special Contributor

Follow Sue on Twitter @suemarek

Subscribe and receive the latest news from the industry.
Join 62,000+ members. Yes it's completely free.

You May Also Like