Industry Voices—Jarich: Was CBRS born too soon?


I’ve come to think about T-Mobile like the old friend so many of us have. The kind who often states the obvious—sometimes positions it as a revelation—but then actually acts on it. Meditating is supposed to be good for you (duh), and I’ve started doing it regularly (wow, nice). People would actually rather have unlimited mobile data plans and unlimited mobile video (you don’t say), and we’re giving it to them (as long as your network can handle it, good for you). Profanity can liven up boring corporate presentations (sure), now go listen to our latest financial results (more interesting than just reading them, I suppose).

So, when I saw the news reports that T-Mobile was advocating for “5G Friendly” licensing in the CBRS band, I wasn’t surprised; increasingly, I’d been having conversations positioning the 3.5 GHz spectrum as a 5G band. Turns out, though, that while T-Mobile is concerned with licensing details, I’ve got a much bigger concern on my mind: whether we’ve actually begun the CBRS experiment too early.

T-Mobile’s argument is fairly straightforward. CBRS is positioned as a shared spectrum band, with protection for incumbent (government) users and two tiers of other licenses: Priority Access License (PAL) users, who will get protection over General Authorized Access users. From T-Mobile’s perspective, then, there are three ways to make investment in CBRS more attractive: longer terms for PALs (10 years vs. three years), larger licensed areas (partial economic areas vs. census tracts) and awarding PALs even if there’s only one application in a market. In its own words, this should “facilitate investment in the 3.5 GHz band and ensure that the United States retains its leadership position in the development of 5G technologies across all spectrum bands.”

And yet, this all leaves a fundamental question unasked. Vendors, operators and industry bodies like the CBRS Alliance are talking up near-term momentum (mobile services next year) based on deployment of LTE in the band. But, should we even be talking about deploying LTE in the CBRS band? 

This isn’t a question for the FCC; it won’t prescribe a technology for the band. For anyone involved in driving commercial CBRS operations, it’s clear why they’d be eager to avoid any delay. The earlier we get started on building out CBRS services, the sooner we can expect to see the ecosystem development and cost efficiencies—supporting new service launches, in turn. For any startup in the space, delays could even be deadly as funding runs out. And, no matter who you are, money today is better than money tomorrow. Unfortunately, moving on an LTE-based CBRS model instead of waiting for 5G is problematic for a number of reasons.

  • Latest and greatest: What we all want. When car companies market their new models, they play up all the great technology. Why? Investment protection. At a consumer level, nobody wants to make a major investment in subpar or soon-to-be-surpassed technologies. The same holds for wireless networks, where the expected life span of investments could be 10+ years. LTE will have a long life, but if something newer and shinier (with untold potential going forward) is just around the bend, wouldn’t you want to start there?
  • Digital industries: The real 5G story. Beyond the core use cases that nearly every presentation on 5G feels obliged to rehash, a core 5G value proposition revolves around new enterprise and vertical market opportunities under the banner of “digital enablement.” Funny thing is, that’s a focus for CBRS too, drawing on the notion that access to spectrum will help drive digital innovation and usage. Does this mean you can’t have one without the other? No. Does it suggest a natural alignment which could help drive 5G forward and usage of the CBRS spectrum? Yes.
  • Global scale: We are not alone. In the grand scheme of things, we are in a fairly important era for spectrum in and around the 3.5 GHz band. In the U.S., we have impending CBRS commercialization. In China, the band (3.3 GHz to 3.6 GHz) is being considered for 5G. Japan, meanwhile, is looking to free up spectrum for 5G, including spectrum between 3.6 GHz and 4.2 GHz. Korea is allocating spectrum for 5G in the 3.4 GHz to 3.7 GHz range. In Europe, there’s a regulatory push to harmonize around 3.4 GHz to 3.8 GHz. I could go on, but you get the idea (and Qualcomm’s Dean Brenner does a better job with a summary here). Much of the world is looking at the spectrum around 3.5 GHz as a 5G band with clear implications for scale, infrastructure costs, device costs and ecosystem innovation.

I completely understand the objections to tightly coupling CBRS to 5G RAN development. We need to begin proving out the shared spectrum scheme. What’s more, this isn’t an either-or dynamic; we will have multimode devices that support LTE alongside 5G NR, and even operators that begin with deploying LTE in the CBRS band can move on 5G at some point in the future.

I also know that inertia applies to network deployments (it’s difficult to think of making a massive push on 5G NR once LTE gets deployed, at scale, in the band) and that there’s a lot at stake with this spectrum-licensing experiment; if sharing proves successful in the U.S. at 3.5 GHz, it could pave the way for similar efforts in other countries and other bands. Handicapping CBRS by starting out with a focus that’s not aligned with 5G R&D road maps or spectrum policy elsewhere may seem attractive in the near term, but the expediency could prove costly.

Peter Jarich is the Chief Analyst (Global Telecom and IT) for GlobalData. Follow him on Twitter: @pnjarich.