Sure, but running it down the street is comparatively a much bigger cost. And I'm not sure the cost of maintaining a radio on every street corner, individual antennas on every house, and maintaining a clear line of site to every individual antenna is going to be cheaper than burying a few extra feet of fiber in the long run.
Plus running fiber from the street to the house means the residents don't have to worry about heavy rain, trees in their neighbor's yard, or buses impacting their internet service.
> Sure, but running it down the street is comparatively a much bigger cost.
The branch factor kills you. I had fiber installed at my house. It took half a day to run it down the main road to the subdivision another half a day to run it to the pole next to my house. Another half a day to trench it under my driveway and install the CPE. If they just ran it to the subdivision and put up a 5G base station, they could serve a couple of hundred users. It would cost far more time to actually pull the fiber to each of those same couple of hundred users.
One tower for a 200 house subdivision isn't going to work.
If you want enough bandwidth to replace fiber to the premises, you're going to use millimeter wave (if you're talking one tower per neighborhood you're already talking about mm wave), so you'd need completely unobstructed line of site to every single house.
The pole would need to be high enough that tower would he a more accurate description, you'd have to keep removing trees and cutting branches to maintain that clear line of site, and rain would cause problems.
I've set up microwave transmitters before for a similar use case. At these frequencies it's going to be much harder.
My guess is that they're going to have to run fiber to the end of the hypothetical subdivision anyway and set up multiple towers throughout.
Also you don't need to actually run fiber to every house to get competitive speeds to 5G. And you can run fiber without burying it. I don't think 5G towers in every suburban neighborhood is going to be a viable replacement for cable.
On my block, they installed about 4 poles. The fiber splicing/install work took about 3 hours. Those poles will serve something like 150 households someday. This is a mid-density urban block with 1-4 family buildings.
It took Verizon two weeks to install external FIOS CPE equipment in my brothers’ subdivision, and probably yielded fewer homes.
That’s a lot of capital cost, and these telcos are looking for margins closer to cellular. They get that by eliminating these costs and getting better regulatory terms.
You're comparing a subdivision to an urban block, and you're not including time and cost to clear LOS obstacles. It's also very unlikely that 4 small poles will have direct line of site to 150 households outside of high density apartment buildings where running fiber directly to the building is already easy.
Surely they'd always install the radios above bus/lorry height? Even in low sprawling areas of the U.S. there's sufficient building height available for that?
I specifically meant when it comes to mounting it on the office/house endpoint. Everything else in between should have another flexibility when it comes to choosing the location.
Yes, both ends the tower and endpoint can be mounted higher than a bus, but because most building are single story, they can't be mounted that much higher.
So a small increases in elevation in an area between the 2 points means that the LOS can still be crossed by a bus.
Plus running fiber from the street to the house means the residents don't have to worry about heavy rain, trees in their neighbor's yard, or buses impacting their internet service.