Mobile ops and Wi-Fi set to scrap for spectrum in the glorious 5G future As the world's boffins prepare to carve up the airwaves again next year, the mobile industry has stepped up the lobbying war over spectrum, fearing a crushing 5G disappointment. Expectations for 5G are now stratospheric, with talk of it threatening the business case for fixed-line retail broadband and Wi-Fi. And maybe even …
Every day I see something talking about stuff that will never happen, like 5G making wifi obsolete, or leading to the creation of whole new ecosystems of devices - nevermind that these imaginary devices could work just well with LTE today rather than waiting for 5G unless they require millisecond latency.
You hear that self driving cars will require 5G for coordinating with other vehicles - as if communicating with a tower, out to the cloud somewhere, and then back again is somehow better than communicating directly with the cars near you. Nevermind that a requirement for 5G would be a problem for cars in much of the US for many years to come...
Mobile operators are making these claims because they want governments to give them swathes of spectrum for free, hoping that lawmakers will fear not doing so will leave their country permanently behind others and become another North Korea.
"as if communicating with a tower, out to the cloud somewhere, and then back again is somehow better than communicating directly with the cars near you. "
I think what they're talking about is that the 5G protocol allows for point-to-point communications, so that cars (or any device) can talk directly with each other without going through a provider.
But early mobile data network transmit moderate bandwidth reliable data signals to allow eg aeroplanes, to inform each other where they are.
Cars will need a massively high bandwidth signal that doesn't work in rain or fog and suffers from lots of complex multipaths when used around a bunch of metal cars
"Cars need massive amounts of bandwidth". Just low latency / high speed
"Complex multi paths" shouldn't be a problem either, as the data has to travel only very small distances
"bunch of metal cars" : by the time 5G is realised, many cars will be composite, not metal
Yeah I'd like someone to explain exactly what need cars will have to pass hundreds of megabytes between each other. All they need to know is where the other cars are, how fast they are going, what they are planning on doing as far as turning/changing lanes/etc. and what "stuff" the car senses around it such as other cars, obstacles, signs, etc.
Even 3G would be massive overkill for this.
I don't think so. The very thought of using a cell protocol for LAN purposes makes me queasy. It would be too easy to have data leak outside of my walls and onto the cell network.
I have to admit, though, that I find 5G uninteresting. It doesn't seem to offer anything that I want that 4G (or even 3G, if I'm not in a hurry) doesn't already provide.
5G was never meant to increase spectral efficiency over LTE, and doesn't. It is just as good as 150 Mbps LTE in that regard. Both max out at QAM 256, though the most recent 3GPP standard allows for QAM 1024 for both LTE and 5G - but you'd practically have to be under the tower to get sufficient SNR for that. 5G has one and only one advantage inherent to the protocol that makes it better than LTE - it offers significantly improved latency.
All the headline hype about 5G being "faster" (as if any phone needs something faster than the 2 Gbps we are already up to with LTE) is because new high frequency bands are being opened up for 5G. Meaning it will increase potential max speeds the same way that LTE > 150 Mbps has increased speed - by using more spectrum at once. There's a lot of room to grab really big swathes of spectrum at 28 GHz, after all.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
