The more operators are confined to being access players, the more they have to worry about the cost of supporting connections. We all have to worry about that, in fact, because the division of the Internet ecosystem into plumbing and gold-plated toilet seats (so to speak) means that there’s a risk that the value we see in the latter will be compromised by failures in the former. Endless demand for capacity can’t be fulfilled at zero profit, so we can expect broadband expansion (geographically and capacity-wise) to be limited by cost of deployment.
Fiber to the home isn’t going to work everywhere. Fiber pass costs have fallen, but they’re still at least triple the pass cost of CATV, and that isn’t going to work everywhere either. The problem with any sort of physical media is the need to acquire right of way and trench the media, then maintain it. We know that urban and the denser suburbs can be served profitably with fiber or CATV, but deeper suburbs and rural areas simply don’t have the demand density needed. One thing that would help would be a strategy that didn’t impose pass costs, in the current sense, at all.
5G millimeter-wave technology, when used in conjunction with fiber-to-the-node (FTTN) offers gigabit speeds and the potential for addressing at least the deep-suburban piece of the digital divide. Stick a node in a suburban area and you can expect to offer high-quality broadband for a mile around it. There’s no need to prep each home or trench media; you just send your new customers an antenna and instructions. The problem is that mm-wave doesn’t penetrate obstructions well, and there have been reports that even trees will create a barrier to service. Some operators have told me that they’re looking at ways to make 5G/FTTH work better, and maybe even reach its potential.
In typical mm-wave deployments, you stick a node and transceiver in a hopefully high location, often an existing cell tower, and with that serve an area roughly a mile in radius. According to operators, getting the transceiver antenna high is helpful because if the antenna is well above nearby trees, it’s only the trees in the yards of customers that are likely to pose a barrier to the service. The problem is that those trees are barrier enough.
Let’s say we can get our transceiver antenna a hundred feet in the air. If you work through the geometry, you find that at a mile range, the line of sight would be at an angle of 1.085 degrees to the horizontal. A twenty-foot tree at the back end of a lot, say 50 feet from the home antenna, would cover an angle of over 20 degrees, which means that it would be in the path of our millimeter waves. To get clear line of sight above that tree, you’d need a tower a couple thousand feet high.
My operator friends tell me that they’ve determined that it would be difficult to make this sort of 5G/FTTN work in wooded areas unless there was a natural high point of considerable elevation. However, there might be another model that could work well. That model could be called “fiber-to-the-intersection” (FTTI).
Look at a typical crossroads, of which there are millions worldwide. You can typically see quite a distance down both streets, in both directions. There are usually trees, but they’re not usually closely spaced, even in suburban areas. The buildings tend to have a fairly standard setback, too, so they line up well. Imagine a millimeter-wave antenna in these intersections; it would have a pretty clear line of sight to structures along the street in all directions.
You may be looking to empower users and businesses with broadband, but what you’re really doing in any rational broadband strategy is empowering buildings, and buildings are most likely strung along roads/streets. Focusing 5G/FTTN on FTTI missions would make sense, then, in multiple ways.
Another point operators made was that arbitrary locations at the best “geographic” point for a millimeter-wave node could well create difficulties feeding the node with fiber. There are always rights of way along transportation paths, but rarely across people’s yards or fields. Without a feed for fiber and nodal power, millimeter-wave is about as useful as an unplugged microwave oven.
There are downsides to FTTI, of course. On the technical side, operators say that the likely number of customers you could expect a given node to support is lower because most streets/roads don’t run straight for a full mile, and curves would introduce barriers, particularly to antennas that had to be at most at the top of a pole. However, there are practical limits caused by terrain and foliage in any millimeter-wave approach, and it’s not clear that FTTI would be worse. In fact, as I’ve noted, operators seem to think it could be better.
There’s also a political issue. A millimeter-wave node stuck on a cell tower isn’t an in-your-face installation. Adding one to an intersection, a place where residents and workers drive through daily, invites pushback. Local government intervention can be time-consuming and costly for operators, and if specific legislation is involved there’s always the risk of a change in administration (at the local, state, or federal level) could swing the rules against any accommodations previously negotiated.
The problem, of course, is whether there’s an alternative. Estimates of just how much capacity a home or business needs to be considered “broadband empowered” vary considerably. Most operators think that 100 Mbps download and 50 Mbps upload would be a reasonable goal. Neither copper loop nor satellite technology can currently meet that standard. 5G cellular, millimeter-wave, and fiber (at least to the curb) are all suitable, under at least some situations.
Some operators (including, obviously, most cable providers) see a combination of fiber and CATV cable as the answer. After all, we deliver broadband and video to millions of locations using that approach. The problem, as even some cable operators will admit, is that it’s becoming more and more difficult to deploy new CATV plant as the demand density of the unserved and underserved drops, which it does as you pick all the accessible apples of demand pockets.
Any physical-media approach to broadband is limited where demand density is low. That means that one of the 5G models (cellular or millimeter-wave) would be a preferred addition to the current model of CATV and fiber. The FTTI interest I’m hearing about represents an attempt by millimeter-wave advocates to deal with the barriers to deployment of their favorite approach.
Most of the operators, including the FTTI and millimeter-wave advocates, would admit that broader 5G cellular usage in home broadband would likely be a better approach. One reason that there’s considerable operator interest in 5G to start with isn’t the fact that you could give a mobile user a couple hundred megabits per second, but that you might be able to give that capacity to a home user. Samsung’s recent achievement of 8Gbps 5G delivery using massive MIMO doesn’t mean that 8G smartphone services are likely profitable, but that it’s possible to support a higher bandwidth for a 5G cell site, enabling that site to deliver home broadband as well as cellular 5G service.
Operators tell me that a pure cellular-5G model to support both home broadband and mobile services isn’t efficient in higher demand density areas, and that millimeter-wave 5G isn’t effective in very low density areas. It looks like operator planners are jiggling their strategies to find the best way to use millimeter wave, to minimize any empowerment gaps they face, and to keep broadband improving and profitable at the same time.