Verizon Communications Inc.
uses artificial intelligence models to help place thousands of 5G wireless transmitters for peak performance.
Later this year, the company will begin a multibillion-dollar deployment of mid-range spectrum, which extends the coverage of its existing 5G ultra-broadband wireless service.
Maximizing coverage with the fewest issuers is a priority, said Shankar Arumugavelu, senior vice president and global chief information officer at Verizon.
“When we build these networks, they are very capital intensive,” he said. “We have to make sure that we are very judicious in how we invest our capital. “
The models, designed by in-house data scientists and other employees, take into account a number of variables that can alter the strength of 5G signals, such as buildings, bridges, terrain, transmitter position, as well as other transmitters nearby.
Verizon, along with its rivals AT&T Inc.
and T-Mobile US Inc., is striving to develop nationwide 5G service, a years-long effort hampered by the lack of available waves for fast transmission and long signal ranges, and by the deployment new network equipment, analysts said.
America’s largest operator, Verizon, started using AI models in 2018 to determine the best places to install transmitters for its ultra-broadband spectrum and is once again using AI for its mid-band spectrum, considered the ideal point in terms of range and speed. Verizon plans to cover 100 million people with mid-range spectrum by March 2022, according to the company.
“Without AI, we wouldn’t be able to do this. Period, ”said Mr. Arumugavelu.
In a U.S. government on-air auction earlier this year, Verizon landed more than half of the mid-band spectrum.
The carrier will pay more than $ 50 billion for these airwaves and related expenses and plans a total of $ 10 billion in capital spending over the next three years to expand its bandwidth range, the WSJ previously reported.
Other operators are also using AI in their 5G plans.
T-Mobile, which is also releasing its mid-range 5G network, has used AI models to help engineers prioritize locations for upgrading its 4G equipment to 5G, said Grant Castle, vice president engineering and device technology laboratories.
The so-called small cell 5G transmitters, or antennas, are roughly the size of a backpack and are placed on vertical structures like utility poles or buildings.
Verizon’s AI system can indicate the best of those perches by weighing various factors in specific zip codes and how well the signal could cover the area, Mr. Arumugavelu said. Using computer vision technology, models can analyze spatial and geographic features (bridges, trees, and building heights around specific intersections, for example) and find the most viable location for a 5G transmitter, a- he added.
The technology can also recommend the best placement for the antennas so that the signal covers the area without causing interference with other antennas nearby, he said.
Billions of dollars are at stake in deploying mid-range 5G spectrum, which is why Verizon relies on AI, said Sid Nag, vice president of research for cloud and advanced computing at technology research firm Gartner. Inc.
“They have to be very precise and clinical to identify where they are going to place the knots,” he said.
Companies are expected to adopt 5G faster than consumers, in part because of the technology’s potential to enable connectivity for drones, autonomous vehicles and remote healthcare, Nag said.
Verizon and other carriers are also making sure that 5G nodes are in the right places because they want their customers to be as happy as possible with the service, he said.
“Businesses are hungry to use 5G,” he said.
Write to Sara Castellanos at [email protected]
Copyright © 2021 Dow Jones & Company, Inc. All rights reserved. 87990cbe856818d5eddac44c7b1cdeb8