5G vs. FTTH: The End of Wire

I’m seeing a lot of articles pitting 5G wireless against Fiber to the Home, some good and some bad. The good ones highlight the fact that carriers are testing new service offerings in urban markets that provide gigabit speeds over wireless; this is real. The not-so-good ones are strawman reasoning that argue against the proposition that 5G will replace DSL and cable for rural broadband services. This is not so good. This piece explains when 5G is a real substitute for wired broadband and when we shouldn’t even be considering it.

The Reality of 5G

Every new technology rolls into the market on the wings of over-hype, and 5G is no exception. We won’t even have real, standards-based 5G networks until 2020, when the standards are done. But there’s already a lot of stuff on the market that claims to conform to this still-undefined standard. This is bad because it dilutes the value of the standard and poisons the reputation of 5G networks before we even see them in action.

What have right now in 5G is a whole lot of trials. Companies with skin in the game – equipment manufacturers and carriers – are testing different technical systems that look good in simulations and under lab conditions to see how well they perform in real-world conditions. This doesn’t always happen with wireless technologies. With Wi-Fi in particular most testing is done in simulations and noiseless chambers.

Real-world testing isn’t as productive as you might think because there are so many variations in the real world: particular sources of noise and interference which may be present in one setting but not in others. That being said, it’s easier to test an indoor system than an outdoor system and most Wi-Fi installations are indoors. Where Wi-Fi really gets hairy is in public spaces with a lot of coming and going such as stadiums and airports, and in apartment buildings. But 3GPP standards like 4G and 5G have to work outdoors, where there are many more variables.

Practical Keys to 5G Success

In a practical sense, 5G only needs to provide each user with 25 Mbps or so to be a realistic alternative to VDSL and cable modem for small families and singles. 100 Mbps with today’s typical usage patterns and you seal the deal for everyone. Gigabit speeds will come to most urban homes someday, but it remains to be seen of they’ll be good for a lot of apps that won’t work just as well at 100 – 300 Mbps.

And 5G is going to be more symmetrical than current wireline systems (other than fiber) but that’s not necessarily very important. The two applications that benefit from high capacity uploads are backups and security videos, but watching four streams at a time is easily doable with 50 Mbps.

While 5G will have advanced signal processing, the key to its success, more than anything, will be lots of spectrum. 5G can use higher frequencies that are less populated today than the low bands some people used to call “beachfront”. When these high bands become more occupied, the signal processing kicks in and we can share more efficiently – with zero sharing loss in some cases.

So there’s no serious doubt 5G will work. The wild card is the willingness of carriers to deploy the small cells that will be required to make it work, which comes down the willingness of consumers to pay for the service.

The Fiction of 5G

5G is an urban solution that doesn’t do much for hooking up rural people to the Internet. So articles that compare and contrast 5G with FTTH are about urban settings. The technology of the present and near future in rural America is fixed location LTE. This is a 50 Mbps technology today with an quick upgrade path to 100 Mbps.

The magic of rural LTE is that carriers go into the market with even more spectrum than they have in urban markets and many fewer people to serve. They have more spectrum available because there aren’t as many active TV channels and not as much division between competitors.

But there will be 5G applications in rural markets, probably smart farms that find 5G and 5G-ish systems such as LTE-U better for outdoor systems than Wi-Fi. I’m suggesting that farms of 1000 acres or more may be better off with privately-owned 5G networks for Internet of Farms applications than with Wi-Fi. The reasoning for this is complicated, but it comes down to reliability, coverage and stability.

Don’t be Fooled

The limitation that both Fixed LTE and FTTH have in rural settings is backhaul cost. It’s expensive to run long strands of fiber along the right of way, even though farmers are pretty good at digging trenches. The reason is that installation is just the beginning of the battle: it’s not enough to install a network, you also have to keep it operating in all kinds of weather. So wireless backhaul will be used in many places for a long time in rural areas.

And if you’re going to rely on wireless backhaul, there’s not much reason not to use wireless for the last three to five miles as well. And face it, wires break but radio waves constantly renew themselves like self-healing robots.

So you can count on rural networks being built out of a variety of wireless technologies in the present and in the near future. As needs for capacity increase, signal processing systems that enable more efficient sharing will keep rural folks ahead of the tsunami.