X

Connect (X)

Tag Archives: iGR Market Research

Why 5G Densification Starts Now

By Iain Gillott

In order to deliver more bandwidth to each user, 5G architectures depend on using more spectrum and more cell densification. More cells in a given area means that more users can be supported than with 4G networks. This is why most 5G discussions include ‘densification’ and the need for many more small cells.

However, in the U.S., at present, this poses a problem. Deploying small cells has continued to prove problematic, mainly due to the zoning and planning issues. With multiple jurisdictions to deal with, mobile operators and others trying to locate small cells have run into multiple roadblocks. While some potential fixes at state and federal levels are in the works, problems persist.

This is not to suggest that no small cells are being deployed; they are, as our research numbers and forecasts show. The problem is that there are not enough small cells to support the densification needs of 5G. While estimates vary depending on who you talk to, iGR estimates that 5G will eventually need about 10 times as many outdoor small cells as have been currently built. Think hundreds of thousands, not tens of thousands.

At present, it is taking about two years (sometimes less but that is a good average) to get permission to build small cells. The actual build takes a few days. So unless things change quickly and the time to get permission to build reduces significantly, if mobile operators want a significant number of 5G small cells in 2019 and 2020, the process needs to start now.

In conversations with mobile operators the belief is that, assuming the current LTE small cells can be upgraded without additional zoning permission, there are sufficient LTE small cells at present to support the initial 5G deployments. The problem comes after this initial ‘inventory’ is exhausted and greenfield 5G small cells are required.

Could the problems with zoning small cells delay or significantly impact 5G network builds? Yes, they could. But likely not until late 2020 or 2021 depending on the jurisdiction. Some cities and towns are allowing operators to build small cells, and these locations will therefore likely benefit from 5G sooner. But those cities that block small cells will, as a result, have to wait to get the full benefit from 5G.

The Current State of the 5G Hype Cycle

By Iain Gillott, contributing editor

Good things come to those who wait, and the wait may be a little longer than the hype about 5G wireless communications would have you believe.


Every new generation of mobile technology comes with its own hype and over-promises — that is simply the nature of the industry. Back in the 2G wireless communications days, the industry was battling over which was best: GSM or CDMA. Millions, and maybe billions, of dollars went into marketing the two technologies, with vendors in distinct camps.

Then along came 3G wireless communications. (Actually, this was the initial generation to be labeled with a G — 2G and 1G were retrospective). A battle ensued between UMTS and CDMA2000/EVDO, and the same pattern followed: Each side promised peak performance, efficient deployment and cost benefits for mobile operators. Proponents of each technology family argued that the other was not real 3G. And devices in this era were mainly flip phones, Blackberries (remember those?) and, at the very end, the Apple iPhone.

4G wireless communications saw the first standardization among the mobile operators with the selection of Long Term Evolution (LTE) modulation of the radio wave that carries the wireless signal. The debates started again as to whether LTE was real 4G and whether another technology, such as WiMAX, should have been selected. But the world went to LTE and ensured that the Long Term Evolution moniker was jokingly referred to as Long Term Employment. The introduction of LTE coincided with the growth of the shiny glass block smartphone designs, and the two grew together — LTE needed smartphones to realize the benefits from data speeds, and the new smartphones needed LTE to make the applications useful and valuable. As smartphone use grew, so did mobile data demand, which resulted in higher network investment for the mobile operators.

Now we get to 5G wireless communications and again, the industry is promising the world: broadband data service closer to a fiber experience; ultra-low latency; lower operating costs; home broadband services; millimeter-wave devices; IoT support on a massive scale; a thinner, richer and better looking you. You get the idea.

In its haste to get to 5G as fast as possible, the initial 5G standard was split into two parts: Release 15 New Radio (NR) to define the radio component and Release 16 to define the 5G Core. To get the entire set of 5G capabilities (broadband mobile data and low latency), you need both the radio component and the 5G core piece. Rel. 15 was finalized in June 2018, and although the 5G Core has been defined, Rel. 16 will not be frozen until the fourth quarter of 2019 or the start of 2020.

And of course, operators have been rushing to be first to launch 5G. Verizon Wireless has launched its nonstandard 5G fixed wireless service in a few markets. Meanwhile, AT&T Mobility promised initial 5G mobile service before the end of 2018.

So, where are we, really, with 5G today? What can consumers really expect? And how close are we to offering a real 5G service?

1. The Verizon Fixed Wireless service is nonstandard, and anyone who buys this today will get new customer premises equipment (CPE) as Verizon moves the initial builds to 5G NR. Why did the company not wait? Because it wanted to be first.

2. The Verizon 5G Home service is available in highly limited areas. Do a search and read some of the Reddit posts — they are enlightening. There are consumers who live on the same street in the initial launch markets, and people at one end of the street can get the service while people at the other end cannot. The initial coverage is highly limited, probably because of No. 1 above.

3. The vendors continue to work on standards (they are always working on the next version), but they have found some 23 problems with the June Rel. 15 freeze that are not backward compatible. In other words, there are problems that need addressing that may mean that devices using the June 2018 freeze will need to have firmware updates or be replaced (to be determined). How did this occur? Simply in the rush to get the 5G NR standard approved, a few things slipped through.

4. So now the question for the initial 5G NR mobile operators is, do they launch with the June 2018 freeze version and risk having to replace devices? Or do they wait a few months or a quarter and miss launching in 2018?

5. Consumers need 5G devices to buy so they can use the new services. AT&T will launch mobile 5G with a puck and not a smartphone or tablet. 5G smartphones will not come until the first part of 2019, with more mainstream devices launching later in 2019. The Intel 5G chipset, which promises great performance, will be available in the second half of 2019, with the first devices using it in 2020. This means that, unless there is a shift in the silicon provider, Apple will not have a 5G iPhone until 2020.

6. The 5G core is needed for low-latency services and a range of 5G capabilities. The major operators are now talking with vendors about the 5G core for deployment starting late 2020. Although this sounds a long way off, remember it is really only 13 or 15 months away — and this is equivalent to next week in network terms. But this does mean that many 5G features and capabilities will not be here until late 2020 or 2021.

7. So for 2019, the talk will be about 5G launches and introductions, but the reality will be more Gigabit LTE devices (with impressive speeds), slowly increasing coverage for the fixed wireless options and 5G devices coming later in the year — for some.

The industry has overreached and overpromised this time. Consumers are generally aware of 5G and may expect to buy services and devices soon. So, while LTE continues to go from strength to strength (Gigabit LTE, for example), 5G needs time. It is a big job to deploy a new network technology, and it really cannot be rushed. We are most likely to look back in half a decade or so and conclude that the initial 5G launch was a bust and we should have been more patient. And those operators that rushed to be first may realize that this is a marathon, not a sprint.

5G will not die. 5G will succeed, but it is going to take time and patience. Good things come to those who wait.


Iain Gillott is the founder and president of iGR, a market strategy consultancy focused on the wireless and mobile communications industry. The company researches and analyzes the effect new wireless and mobile technologies will have on the industry, on vendors’ competitive positioning and on its clients’ strategic business plans. Visit  www.igr-inc.com.

This article originally ran in the March 2019 issue of AGL Magazine.

OpEx to Overtake CapEx by 2015 — U.S. LTE Forecast

As network build outs begin to taper off and subscribers take up LTE devices, carrier operating expenditures will surpass capital expenditures by 2015 according to a report released by iGR Market Research last week.

“Most of the largest U.S. operators are already well advanced in their LTE network rollouts,” Iain Gilliott, president and founder of iGR, wrote in a press release. “Others may lag behind, but LTE deployments overall are still progressing quickly – perhaps more quickly than could have been foreseen 12 to 18 months ago.”

The major operators have been investing capital expenditures in LTE over the last two to three years and now as their LTE subscriber base is growing, they are incurring increased LTE operating expenses, according to the report.

Based on the anticipated rapid growth of LTE subscribers and data traffic on the networks, total U.S. LTE infrastructure OpEx projected to be $57.4 billion, while CapEx is expected to be only $37.7 billion between 2012 and 2017.

“Operators are striving to provide sufficient coverage to be competitive and sufficient capacity to meet the needs of the growing subscriber base, while minimizing unnecessary CapEx and OpEx,” said Gillott. “This report clearly shows there remains a limited amount of time for significant CapEx expenditures in the U.S. LTE infrastructure market, and that the operators’ LTE OpEx levels will overtake LTE CapEx levels in two years.”