Researchers at NYU Wireless are warning the cellular industry that it needs to change the "alpha-beta-gamma" (ABG) model used in the past and start using simpler, more accurate models for predicting signal coverage for frequencies above 6 GHz.
A new study by Theodore (Ted) S. Rappaport, the Ernst Weber/David Lee professor of electrical engineering at NYU Tandon School of Engineering and founding director of NYU Wireless, suggests that the three-parameter ABG model used in the past by 3GPP for predicting signal coverage might spell trouble at frequencies above 6 GHz. The 3GPP RAN group is expected to discuss channeling at an upcoming meeting in Nanjing, South Korea.
The new research, published this month in the IEEE Vehicular Technology Transactions, is based on an extensive set of 30 measurement campaigns from 2 GHz to 73 GHz. It was conducted with co-authors from Aalborg University in Denmark and industry partners and is part of the Ph.D. work of NYU Tandon School of Engineering graduate student Shu Sun.
The study argues for a better, simpler, alternative model that uses just a single parameter, the "path loss exponent," which can be used as a global standard for predicting signal coverage.
The NYU researchers did not always believe that a new channel model standard was necessary. A few years ago, "we thought the 3GPP model was better," Rappaport explained to FierceWirelessTech. But they looked over the data they've collected over the last few years and realized the one-meter free space reference distance model had the same accuracy, yet the model only needed one parameter -- the path loss exponent -- causing them to question the older model.
They did a sensitivity analysis, and that's where "we were shocked to find that the simple CI (close in) model was not only simpler," but it's also much more stable and accurate, he said. "We think it's really important for 3GPP because it shows how the fundamental physics, the frequency-dependence in the first meter, provides a much simpler, easier, more intuitive model" that can be used over these frequencies. The ABG model, which has been used for decades, doesn't do that – it's non-physics based and "you can't really stake any confidence on it."
Channel models are kind of like accounting, he said. Every carrier and manufacturer develops their own proprietary channel model that they use, similar to how a corporation might keep one set of books that they show the public and another secret set of books that shows what's really going on.
While the 3GPP is working to develop a public standard channel model that everyone uses for comparison and future planning, "I'm arguing for a propagation model that works for everything below 6 Gigahertz and everything above 6 Gigahertz, all at the same time," and it's fundamentally more accurate and more stable than what's being used now.
Rappaport said he is certain that by using the old ABG model, carriers and manufacturers will greatly underestimate the number of base stations or antennae, for example, that they will need in 5G networks. The models used today for non-line-of-sight (obstructed signal paths) estimate that the signal level within the first 20 or 30 meters of the transmitter -- of the base station -- will be stronger than even in free space. "That's ludicrous," he said. "The signals are going to be much weaker."
Dino Flore, chairman of the 3GPP RAN group, told FierceWirelessTech that the RAN1 group will handle the NYU Wireless communication along with a host of other issues during a meeting this week.
It's not clear who exactly is arguing for the older model. No one has publicly refuted the results of the NYU research. Many of the industry's biggest players are listed as industrial affiliates of NYU Wireless, including Nokia, Ericsson, Huawei and Samsung. Among the authors of the research paper are engineers from Nokia, Nokia Bell Laboratories and Qualcomm.
Rappaport explained that the one-meter reference distance is based on the fundamental physics of radio propagation going back to the work of Harold Friis at Bell Laboratories in the 1930s. "The use of just a single model parameter, the path loss exponent, and a standard reference distance of one meter, makes the CI model easier to use, more accurate, easier to understand, and more compact in complex computer simulations that will be used to experiment and design future 5G Cellular," he said. "And it works well from 500 megahertz all the way to 100 GHz. The physics should work well into the Terahertz region."
Verizon's Shammo: 5G pilot in 2017 is all about fixed wireless, not mobility
NYU Wireless releases open source 5G channel simulation software
NYU Wireless lobbies for new methods of channel modeling for mmWave