SDX Central
Matt Kapko
March 27, 2019
Millimeter wave (mmWave) spectrum, for better or worse, is going to play a critical role in 5G networks. As such, operators will encounter many challenges as they engineer their networks to reach a level of availability and resilience required to fulfill the promise of 5G.
The problems are varied and complex, but industry analysts remain convinced that operators and the global ecosystem of vendors will eventually overcome mmWave challenges. One of the key sticking points — the pervasiveness of a cellular signal running on mmWave spectrum from a small cell — is causing some to pause after a new report concluded that mmWave propagation rates are lower than initially expected.
Wall Street analysts MoffettNathason conducted an in-depth study of Verizon’s fixed 5G network in Sacramento and determined that each small cell deployed in the market was serving an average of 27 eligible addresses. Although the firm couched its findings as “highly preliminary,” it stands to reason that operators may need to purchase and deploy millions more small cells than anticipated to scale a nationwide 5G network using mmWave spectrum.
Millions of Small Cells
FCC Commissioner Brendan Carr previously estimated there are roughly 300,000 cell sites around the country today, and believes 5G is going to require a 10- to 100-fold increase in cell sites. The MoffettNathanson’s findings suggest mmWave could require even more.
Michael Thelander, president and founder of Signals Research Group, came to a different conclusion after conducting his own on-the-ground research on Verizon’s 5G Home network in Sacramento. Verizon’s online mapping tool “grossly underestimates available coverage. It’s either a flaw or they’re being highly conservative,” he said in a phone interview.
“I was literally testing in front of homes and got great performance and I’d enter that same address onto the website and it would say coverage wasn’t available,” Thelander said. The results of his firm’s tests were “incrementally very encouraging” relative to most peoples’ expectations for millimeter wave.
“The magnitude of the impact is not nearly as severe as people think,” he said. “When you’re dealing with that much spectrum and that much bandwidth, if your signal gets cut in half because it goes through a tree you’re still left with a lot of bandwidth at the other end.”
Verizon Looks Forward to 5G NR
Verizon, which plans to have standards-based mobile 5G network in at least 30 markets by year-end, also takes issue with MoffettNathanson’s conclusions.
“I don’t think some of the assumptions in the report about our ability to scale 5G are accurate given that the testing done was on the 5G Technology Forum (TF) standard,” Kevin King, director of corporate communications at Verizon, wrote in response to questions via email.
The nation’s largest wireless operator has been forthright with its plan to transition to standards-based 5G New Radio (NR) equipment and expects that, along with multi-purpose network deployments and technical enhancements, to lead to improvements.
“The report findings assume 5G technology development freezes right here,” King wrote. “The technology will continue to improve and we will continue to enhance the network.” He added that Verizon is pleased with the 5G Home network performance in its four initial launch cities and said “the performance of the mmWave spectrum is what we expected.”
T-Mobile US CTO Neville Ray reacted positively to the MoffettNathanson report. “We’ve always said nationwide 5G will require all bands — low, mid, and high,” he wrote in an email to SDxCentral. Verizon’s 5G Home service was a “trial of pre-standards 5G TF that could never scale. The MoffettNathanson report tells us what we already know — mmWave can’t scale to nationwide 5G.” T-Mobile US has said it plans to have a nationwide 5G network in 2020.
Ted Rappaport, founding director of NYU Wireless, told SDxCentral it’s important to note these are “very early days” for 5G and mmWave spectrum, and cautioned against using a limited deployment of non-standard technology as a benchmark. “As the global standards of 5G NR are implemented you’ll see vast improvements over that particular study,” he said. Rappaport played a key and early role in demonstrating the potential of mmWave spectrum.
AT&T says it is experiencing better-than-expected propagation on mmWave spectrum with assistance from beam steering and beamforming. “We plan to use a combination of spectrum bands, including lower bands to deploy our nationwide 5G footprint,” an AT&T spokesperson wrote via email. “We’ll use mmWave in denser, urban areas to deliver 5G and low band spectrum in more suburban and rural areas.”
AT&T, which launched its mobile 5G network in 12 cities late last year and says it will have nationwide mobile 5G coverage by early 2020, says it’s achieving month-over-month improvements on its early deployments. “It is true that any application of 5G over mmWave requires small cells to help the signals travel farther, but we haven’t found that it’s taking more small cells than originally anticipated,” the AT&T spokesperson said.
All About the Uplink
“Verizon’s engineers are admittedly still learning the ropes on the deployment. It’s an order of magnitude increase in carrier frequency with completely new kinds of antennas and much wider bandwidth,” Rappaport said. “I’m not worried at all. We did the research seven years ago and we know it’ll work. It’s just a matter of a simple learning curve.”
Rappaport and Thelander agree that the primary limiting factor in 5G on mmWave spectrum is the uplink. That is the connection between a device and a cell site. “That’s a key design parameter in all cellular, so they’re going to have to learn this whole reciprocal design for millimeter wave,” Rappaport said. “We’ll figure it out just like we’ve always figured it out.”
The radius of uplink coverage provided by each 5G small cell is also a moving target, depending on multiple factors. But the baseline for a 5G small cell operating on mmWave spectrum should reach 100 to 200 meters in a dense urban environment, up to 900 meters in suburban areas, and 2 kilometers or more on a tall tower in rural areas, Rappaport explained.
Indeed, most of the challenges that U.S. operators encounter in 5G can be attributed to the limits of mmWave spectrum, Thelander explained. AT&T, T-Mobile US, and Verizon all own licenses throughout the country for mmWave spectrum and plan to use it for 5G deployments, but each carrier is taking a different approach with respect to how mmWave takes the lead or complements additional spectrum holdings.
Sprint, which didn’t respond to a request for comment, doesn’t own many licenses for mmWave spectrum, but it will gain greater access to the technology if its proposed merger with T-Mobile US is approved by regulators. Sprint has said it plans to launch mobile 5G in four cities in May, expanding to a total of nine cities by the end of June. The carrier is relying heavily on its 2.5 GHz “mid-band” spectrum for that launch.
“The operators will be putting 5G into their lower frequencies … but it’s already spectrum that’s being used by LTE so you can’t just turn off the LTE traffic. The amount of bandwidth that’s really available there is somewhat limited,” Thelander said. “For AT&T and Verizon, millimeter wave is kind of what they’re stuck with.”