5G: Three Areas to Watch5G: Three Areas to Watch

Sometimes I wish the 3GPP standards organization would just quit numbering random milestones in the progress of cellular technology -- though they are good for the consulting business. In reality, there are important things going on all the time in wireless, but when 3GPP identifies a new “generation,” all of a sudden everyone has to have an opinion.

Over the past month, everyone from President Trump to CNN have weighed in on 5G, and while there were snippets of fact to be found in each of these reports, they have largely been used to foster some political viewpoint or other. One of the big reasons I chose to work in technology is, we don’t do politics.

Without a doubt, 5G is important, but I look at those “generation” announcements as mere road signs. New developments are introduced into the cellular market continuously, and at a pace that would make your head spin. Some of those developments get advertised like the performance boost from the move from LTE to LTE-Advanced or the incorporation of massive MIMO technology on exiting 4G networks. Some changes like the architectural shift to centralized radio access networks (C-RAN or Cloud RAN) typically don’t leak out to the general public at all because no one outside of the cellular business will likely know they even happened.

Given the ruckus that has grown up around what should probably be classed as “non-news,” we should take a look at where we are with 5G and the things we as enterprise buyers should be focused on as this story evolves.

Initial 5G Performance

The most obvious advance with 5G will be what the 3GPP terms Enhanced Mobile Broadband. That is essentially the same wireless broadband you have today, only faster. The headlines tout a boost in peak data rates from 100 Mbps to 1 Gbps, but issues like signal strength and the number of users sharing your cell or sector have a much greater impact on the boosts actually realized.

The other major developments are support for Massive Machine Type Communications, which means greater device density for applications like Internet of Things (IoT), and Ultra-Reliable and Low-Latency Communications. Regular 5G communications will have lower latency than current 4G, but still measured in 10’s of milliseconds. The new part is that there will also be services with latencies around 1 millisecond for applications like autonomous vehicles. The big thing to know about these last two is that they are “futures,” though I remain fully confident the operators will deliver them in due course. The first thing we expect to see is faster mobile broadband, but even that will take some time.

Currently, there are over 200,000 cell towers in the U.S., all of which will eventually have to be upgraded to 5G. That’s the type of job that takes years to complete. In the meantime, most initial 5G users will encounter the service through small cells, tens of thousands of which will be deployed as part of the rollout.

However, if you promise things are going to get faster, as the operators have, someone is going to step up to test your claims. PC Magazine reported on a Reddit post that a cited a 5G test on AT&T’s network in Haviland, Kan. that showed a 5G download rate of 194.88 Mbps versus a 4G test that showed 187.44 Mbps -- a mere 4% improvement. To his credit, the author points out limitations in the Netgear Nighthawk hotspot that could have been a contributing factor, but the important lesson here is that we are all still way early to be looking for meaningful performance information.

This all reminds me of an experience I had with Rogers Communications some years back when the company was first rolling out 3G. A VP of Engineering at Rogers headquarters was showing me a speed test on his laptop posting some amazing (for their time) download speeds. While I tried to be polite (as I always do), I had to point out that he was one of maybe 10 people in all of Toronto with access to their 3G infrastructure, so I wasn’t exactly sure how meaningful this demonstration was.

The message is, whether the data rate is higher or lower, it’s still way too early to be counting on any real performance improvements. And, if you have an iPhone (none of which will have 5G until maybe 2020), your improvement will most certainly be nil.

What Enterprises Should Be Watching in 5G

While 5G is not yet delivering a life-altering experience, there are some key developments that will have a more immediate impact on enterprise customers. These three areas to watch are:

Continued on next page; click below.

Millimeter Wave Systems

One of the most significant developments in 5G is the ability to operate on a much wider range of radio frequencies than earlier generations of cellular technology. That could result in transmission rates vastly higher than what 4G can provide, but possibly at the cost of reliability.

Prior to 5G, U.S.-based cellular carriers have operated their services on licensed radio channels in the range between 600 MHz and 2600 MHz (that includes Sprint’s 2.5 GHz or “Band 41” channels). The loss characteristics of the lower portions of that (i.e. 600 MHz to 900 MHz) are most attractive for covering large areas.

The overall plan with 5G is to cover the wide area with lower frequency channels (i.e. 600 MHz to 900 MHz.), but operate small cells using other, higher frequencies including what are called millimeter waves, in dense areas underneath that lower frequency wide area umbrella. Of course, all of this is predicated on the assumption that the carriers can use these higher frequencies to deliver reliable service.

The 3GPP has identified some nearby frequency options for 5G systems, including the 3.5 GHz CBRS and 5 GHz Unlicensed NII bands. Those bands, along with the original licensed lower frequency bands are grouped together as “Sub-6 GHz.” The performance characteristics of those bands, particularly the 5 GHz band we now use for Wi-Fi, are all well understood.

However, the portion of the spectrum that is getting the most early interest with 5G is the millimeter bands, a reference to the wavelength of the radio signal (a 10mm wavelength is equivalent to about 30 GHz). In the U.S., the bands that are of immediate interest are 28 GHz, 39 GHz, and 24 GHz; portions of the latter are currently being auctioned by the FCC.

Verizon and AT&T have significant holdings in millimeter bands. Verizon got a major hunk in the 28 GHz and 39 GHz bands through its acquisition of Straight Path Communications. AT&T did the same with its acquisition of FiberTower, though many of those licenses were returned to the FCC after the acquisition.

Operating mobile devices at these frequencies is a very new area. Back in the 1980s, we got the first 23 GHz point-to-point microwave radio systems that operated up to about one mile. Operating an engineered link with directional antennas outdoors is far easier than supporting mobile devices indoors with walls and furniture with which to contend.

Qualcomm has been very active in making this idea a reality, pioneering narrow-beamforming and adaptive beam steering antenna modules that it says can compensate for the problems with millimeter waves. The company describes the challenges and how they approach them in this blog.

These approaches, the company says, will operate up to a couple of hundred meters, or roughly the coverage area of a Wi-Fi access point, so this technology is clearly aimed at small cells. The additional challenge introduced with millimeter waves is the more "linear" nature of the signal (i.e. they don't turn corners very well). That can make for coverage problems in urban canyon environments. From a technical standpoint however, the ability to use these frequencies in a mobile environment is a big deal. If it works effectively in the real world, this makes practical the whole idea of improving coverage with millimeter wave small cells in dense use areas.

A lot of the performance improvements in 5G are predicated based on being able to deliver good performance and reliability over millimeter wave bands, so this is one of the big areas we’re watching.

Indoor Coverage Options

The plan of having a low frequency umbrella with millimeter wave small cells providing reinforcement in dense usage areas is great for outdoor spaces, but we’re still left with the question of how we improve 5G coverage indoors -- particularly for large enterprise locations and office buildings.

Qualcomm claims that its smart antenna millimeter wave technology also works in indoor environments. However, two decades experience with Wi-Fi networks operating at much lower frequencies has taught us that an indoor environment introduces a litany of potential problems. Qualcomm may be right, but we’ll be putting the issue of indoor performance under much higher scrutiny.

For their part, the operators aren’t saying much on their plans for indoor coverage in a 5G environment, but that may be because no one is asking them. From my experience, the level of technical understanding that goes into cellular purchases is pretty low, so more customers may have to experience this problem before they press the operators to get around to dealing with it. Ensuring good indoor coverage for 5G voice users will be important under any circumstance (we’ve got the data problem covered with Wi-Fi), but if the idea of private 5G is going to take hold, the issue of indoor coverage will be paramount.

Private 5G Networks

Solely out of duty to our readers, I have written about the idea of private cellular networks as a potential replacement for our existing private Wi-Fi networks. This is in spite of the fact that I have found little enthusiasm for this idea among any of my clients. However, my focus is primarily on the U.S. market.

Volkswagen recently announced it’s planning a private 5G network for its 122 German factories by 2020, using the 3.7 to 3.8 GHz spectrum that has recently become available in Germany. Reportedly, Siemens and Bosch have already installed 5G test networks in selected plants, so Volkswagen is not alone in this pursuit. German operator Deutsche Telekom told Enterprise IoT Insights that it expects only 20 to 30 enterprises in Germany will likely seek to run their own private cellular networks, but that is far more interest than we have seen in the U.S.

The great thing about cellular is that every national market develops differently. So, if Germany wants to take the wheel on testing private 5G cellular networks, the rest of us will be watching the outcome with great attention.

Conclusion

Wireless is still the greatest business you could possibly be in today. On the surface level, the range of capabilities a mobile user has access to today has run miles ahead of what we could have envisioned 20 years ago, much less when cellular was first introduced in 1983. The part of that not featured in the papers is the unbelievable collection of genius that keeps advancing the forefront of achievement in wireless, routinely making the impossible possible.

Consumers can comfortably take all of this in stride, but enterprise users will have to become more knowledgeable about how these wonders actually work and what challenges they may have to address. We’re “only infrastructure,” but there is now a whole new way of living and working that has come to depend on this.