Lower 3 and 7/8 GHz Bands Present Major Complications: Ghosh
Developing rules for opening the lower 3 GHz band, a top focus of U.S. carriers (see 2404080063), won’t be easy, Monisha Ghosh, engineering professor at the University of Notre Dame and former FCC chief technologist, said during an RCR Wireless virtual test and management forum Tuesday. Much discussion at the forum centered on the challenges of performance testing in evolving 5G networks.
Sign up for a free preview to unlock the rest of this article
Communications Daily is required reading for senior executives at top telecom corporations, law firms, lobbying organizations, associations and government agencies (including the FCC). Join them today!
While 3.1-3.45 GHz is studied, “we don’t know exactly what is going to end up here,” Ghosh said. “All the problems that we are facing in 3 GHz will continue in any kind of mid-band sharing that we look at,” she said. “How do we design cellular systems that operate as well in shared spectrum as they do in exclusively licensed spectrum?” There’s “a lot of work that needs to be done,” she said.
In addition, the 7/8 GHz band will prove complicated because it’s already crowded with federal users, Ghosh said. “How do you share this band?” There’s already sharing between fixed service, fixed satellite service, mobile satellite service and meteorological satellite users, she said. “There’s a lot of work that needs to be done before this band can be brought up for sharing.”
Ghosh predicted sharing will be a major feature of 6G. “Dynamic sharing is going to be increasingly important,” she said. Ghosh is optimistic about sharing's future, given the work of industry and academia. During the next two or three years, “we have to take technologies from the lab, from academic papers, and prove them in the field.”
Spectrum sharing began with TV white spaces more than a decade ago, but today we have already deployed advanced sharing in the citizens broadband radio service band and 6 GHz, Ghosh said. None of the solutions so far is “very dynamic” and most rely on databases, she said.
The goals of 5G, like connection-everywhere and being able to serve numerous users with low-latency connections, are starting “to become a reality” with 5G-advanced, said Emil Olbrich, vice president-network security at Signals Research Group. AI and machine learning is “the big one,” he said. Even simple devices like toasters are starting to use AI, he said.
“It’s tough to standardize innovation,” Olbrich warned. Getting channel state information and being able to do beam management in an open interface, “those are not easy things to do and how do you measure that?” Questions remain about whether network advances will mean an overall improvement in system efficiency or better performance, he said.
Another problem is that much of the focus is on how well a network conforms to the standard, with a minimal amount of performance testing, Olbrich said. As new technologies become part of the network, “we don’t necessarily see the performance that was promised or guaranteed,” he said.
“There’s a lot going on in the network” as carriers try to “monetize” their investments in 5G, said Ross Cassan, senior director-assurance strategy at Spirent Communications. “We see a lot of stress being put on the carriers in terms of being able to implement new features like private 5G networks, fixed wireless, turn on slicing,” he said. All of that just adds to the complexity of the networks, he said.
As a testing and measurement company, Spirent works with carriers to use automation, AI and machine learning to minimize “noise” on the network, “to do things like fault validation,” Cassan said. “Making sure that if we’re going to send an alert to a human or on a dashboard” it’s something that really needs addressing, he said. Automation helps isolate a problem and gives a carrier a clearer view of what’s causing it, he said.
Testing and measurement is critical in 5G, said Mark Watts, Verizon associate fellow-global network and technology. Hardware and software must "be validated and confirmed that it’s acceptable to be in a reliable and relied upon network like Verizon’s,” he said. Also important is network readiness, “whether the network itself can provide the types of services that are expected in 2024 and beyond,” he said. When a carrier rolls out a service, it has to be “validated from a user, end-to-end point of view,” Watts said.
As networks move to the cloud and a software environment, providers need staff with different skills than previously, said Jefferson Wang, Accenture senior managing director. “Sometimes it’s easier to take a traditional network engineer … training them to make the transition into new processes and tools” rather than starting with a software engineer and “then trying to teach the network side of things,” he said.