Samsung sends gigabit '5G' signal TWO WHOLE KILOMETRES The world is getting excited at the advent of “5G” wireless systems with a demonstration of a gigabit air interface using the 28 GHz band by Samsung. It's not too bad as a piece of early-stage academic technology demonstration, as it happens: the engineers used 64 antenna elements in what you might think of as a “massively …
The lower the frequency, the less bandwidth us usually available. So, all this new 700 MHz stuff that is appearing from the old UHF TV channels won't have lots of capacity. They demonstrated 1GHz bit rate at 11 GHz, which is just about right. Assuming 8 bits per signaling element (good luck with that!) you still need 125 MHz of bandwidth to get stuff through. Finding that much RF real estate is VERY difficult at low frequencies as the pie is quite subdivided for all the users.
So, you have high frequencies, and there the path loss starts climbing in and you end up heating the air you pass the waves through. This indicates a higher power in BOTH directions which gets difficult in something hand held.
Bottom line: Good luck, you WILL need it!
The 28GHz band sits squarely in the Ka band, which is widely used in satellite communications. I'm not aware of it being used in-space (though it probably is), it's just the applications where you're blasting microwaves up into space aren't power- or space-constrained anywhere near the amount a phone is, so even 90+% attenuation from rain isn't a massive deal.
Higher frequencies are limited in range due to being absorbed by water. That is not necessarily a Bad Thing. It means that the cell towers can be placed closer together without interfering with eachother as much.
Where higher frequencies break down is that they don't make it through walls etc as well.
The scale height for water vapour in the atmosphere is only about 2km; if you put your satellite dish 1 or 2km above sea level there is little left; hence the placement of radio telescopes for frequencies > 20GHz preferentially in high deserts. Second preference is for cold places where the water vapour has frozen out. The water vapour effect is very little below 10GHz so that is why most TV/GPS/GSM/WiFi stuff is there.
If you transmit strait up (or down) well above sea level you will only get a small atmospheric absorption; if you try this at sea level and then at ground level for many tens of kilometres (e.g in rural areas) it would be a different story. You could get 28GHz system to work well over short distances (like cities or towns with lots of towers)
I agree, although this is mainly a problem with the operators of the spectrum. I have been in countries where you get fantastic, high-speed, 3G coverage. I was even in the basement of a shopping centre with excellent coverage. Mini-cells, microcells, femtocells, etc are the key and UK operators really need to step up.
The 11GHz band Docomo used and the 28 GHz band Samsung used is the range where DBS satellite signals are broadcast. As has been pointed out, water will attenuate these high frequency signals (that's why you will see "rain fade" as a heavy storm moves in, often before a single drop falls) But consider that satellite signals must have CLEAR line of sight. You can't install a dish aimed through a tree, the leaves of even a small tree will massively attenuate your signal.
Granted, the satellite is 22,000 miles further away than the cell tower, so the incoming signal is far far weaker, on the other hand you don't use a half meter dish on your phone as an antenna, either. So not only will getting a signal indoors present problems, even getting a signal standing up a tree could be an issue.
Yeah I know, you can install local boosters all over the place to avoid these problems, but who is going to pay for all that, just to get 5G data rates instead of 4G? The use case for 4G is pretty clear - you can stream video to your phone, and web pages download almost instantly (well, they would if it wasn't for all the back and forth in the protocol, but the actual download is a fraction of a second at 4G speed) What's the use case for 5G? Multiple 4K video streams to a device with a 4-6" screen? Ummm, sure....
The 2km spacing of the towers is also going to be a problem in the US. The reason why AT&T's 3G coverage sucked/sucks in the US was because GSM requires small cells and getting permission to put up towers all over the place has been a huge problem for them. Verizon's CDMA technology had a much larger cell size and thus they were able to cover an area with far fewer towers than AT&T required.
Between the inability to pass through anything and the tiny cells, 5G, at least in the form mentioned in the article, will never come to the US except in very limited areas. The most densely populated areas also have lots of tall buildings. If you're walking down the street and turn the corner you might lose the tower you were communicating with. Or if you step under a cloth awning. Or get behind a streetlight (high frequency signals are VERY directional)
I'm pretty skeptical we'll ever make use of such high frequency bands for mobile communication. Not that this will stop carriers from screaming FIVE G!!!! from the rooftops :)
It's simply a tech demo, mainly to develop the MIMO technology I would guess. I would also guess you would have 64-MIMO at the base station and something more reasonable like 2x2 MIMO in the device. The main point of MIMO in a system like this, your phone will get a good signal from a few of the MIMO elements, someone else's phone might get a good signal off a few others, so if the equipment takes advantage of this the spectrum can be reused with little or no interference. Before WiMax and LTE came out there were all sorts of pie in the sky tech demos too, but they allowed them to develop practical 4G and brag about the process.
and so in effect the spectrum is reused without interference if this is all handled right.
So according to the author no work should be done on future technologies until some undtermined point in the future. For sure this is about patents and control over standards but the work needs to be done (for a given value of need) do we care which company does the work? They're all as bad as each other.
Exactly Mr Spaniel, the aim of the game is to get there first and work out what patients you will need. Not actually make the thing, just get an idea of what would be required for the patient applications.
Its been a long time since anything in the coms world was about innovation or actually making something new
Smaller antenna ;-)
I think we are kidding ourselves if we think the next generation of standards will be donated by some kind university. Right now companies are doing r&d into potential ways to improve. At some point they will get together, a standards group will be formed and a standard will be defined. The companies want control over that, this means owning patents that cover useful technology, and they have to sell them to anyone under FRAND terms, they key is to not find another company owning all the patents for the standard. This is no worse for us than 4g, 3g or 2g. It's basically a patent detente, they will make decent money off the patents based on volume but not exceptional amounts plus they will be able to nudge standards in a direction that suits them but not by much. I just don't understand the fuss about a company doing r&d, surely we should be more concerned about companies with patents who do no r&d??
There are quite alot of these devices around.
They have power, they have elevation, they are more densely clustered in highly populated areas - cover the vast majority of "places where people are"...
Is it just me that thinks they'd make a good network for mini/micro.../atto cells - with a secondary mesh network connecting them together.
A national network of these, wholesaling out peering links (with authentication and subscriber authentication handled by the "ISP"). I know the concept of BT MkII is unattractive, but it has to beat trying to roll out 4 identical networks with high coverage...
Lamposts is a great idea! If people can use them to power their pakalolo farms in their houses their has to be enough juice for a small cell station!
Not sure on the benefits re one network, tower sharing is sensible, an all in network means no competition so if you fancy a higher quality network and are ok with paying more or want a lower quality network but all you can eat data theres no chance to move. The quality of service would be set. Not a huge issue, I'm just not totally convinced, although theres a strong argument for a single shared network in rural areas. That and allowing multimode towers like Sprints in the USA, linking a license to a technology needs to go.
...used 64 antenna elements in what you might think of as a “massively MIMO” system....
Presumably all well and good when all 64 antennae at each end are happily talking to each other in isolation. What happens when umpty-something other bundles of 64 in the area want to chat at the same time?
...how large this 64-element phased array was. Would I, for example, have to be carting around something the size and heft of a galvanised steel bin lid, in order to benefit from this 'advance' in antenna trickery?
Kinda defeats the object of portability, if that's the case, really. *shrug*
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
