Saturday, January 22, 2022

Still Many Unknowns about 5G

Carping about the value of 5G or user experience benefits is not ever hard to find, especially in the United States, where early 5G has had to rely on low-band spectrum (low speeds) and millimeter wave (low coverage) rather than mid-band spectrum, as is the case mostly everywhere else in the world. 


Still, there are some potential long-term changes that have nothing to do with the current state of user experience. 


When a much-hyped new network is under construction, with a full national rollout taking three years or so, impatience is probably inevitable, though 4G seems to have been greeted more favorably, when the primary attraction was speed. 


At least one issue for 5G is that 4G speeds often are quite satisfactory--for the moment--for nearly all the applications people actually want to use on their phones. That makes the siren call of “faster” less useful than in the 3G to 4G transition, when the speed difference was notable since 3G was clearly not fast enough for many apps and use cases. 


Video, for example, was painful on 3G and high-graphics-content web pages also were an issue. 4G fixed those problems pretty quickly. 


It is hard to say the same for early 5G, as low-band speeds are not so different from 4G and millimeter wave access is geographically limited. 


But the 5G rollout also is occurring under unusual circumstances. Because of all the pandemic-related shutdowns, people have not been “out and about” as much as they had before. 


More schooling and work is being conducted from home, which means most users are on Wi-Fi, which also means the benefits of 5G speed would not be so obvious, or so important. 


If, as many expect, the “return to the office” does not return us to pre-Covid levels, there will be a “permanent” reduction of the value of mobile access, simply because there will be less “out and about” needs for connectivity than before. 


We will have to see what happens when mid-band spectrum is universally deployed, and many users--perhaps most--also are on unlimited usage plans. Will the at-home default to Wi-Fi continue at present levels? It is possible it does not. 


Will some percentage of users find “5G all the time” provides a better experience than their home broadband? Possibly, if those users are buying entry-level home broadband service. 


Will price benefits from bundling of mobile and fixed wireless, plus speed that is “good enough” shift demand, and if so, how much?


Will fixed network home broadband providers see share losses? As the U.S. home broadband market is very nearly a zero sum game, it is possible. 


Mobile service provider capex also could be affected. Having historically put more capacity “downtown,” will reduced urban core working time then delay capacity upgrades there, but increase them in suburban areas where people are spending more time?


And how long will it be before new devices for capacity-hungry augmented or virtual reality become popular? How much of that bandwidth will be “5G direct” versus “Wi-Fi” supported? We do not know yet. 


And that is to say nothing of new use cases for network slicing (virtual private networks) that could become important for enterprise users. Edge computing combined with 5G as well as private networks should also become important. But the level of demand is not yet clear. 


The point is that aside from 5G value for consumers, there are other effects for business users that do not necessarily hinge on speed. Latency performance or security could be the key drivers. 


On the network infrastructure front, where capacity needs to be added over the next five years could be different from past patterns. Even if total usage goes up, it might not increase at the cell sites we normally would expect. 


There is much we could yet be surprised about, where it comes to 5G. 


No comments:

Post a Comment

Is Sora an "iPhone Moment?"

Sora is OpenAI’s new cutting-edge and possibly disruptive AI model that can generate realistic videos based on textual descriptions.  Perhap...