For those of you who don't know, here's a brief primer on LTE.
It stands for Long Term Evolution. It's the next step on this mobile phone world journey. There's a particular new technology in it (well, it's in other areas too, but that's for another time) called MIMO, multiple input, multiple output. Simply put, this is having multiple antennas (antennae are on ants, its antennas in physics) in a handset working together to make the data speeds ridiculous. The more antennas, the more ridicluous. If you have a phone with two antennas and a base station with two antennas, there's 4 multiplexed data channels. 3 antennas at each end, there's 9. 4 antennas give 16 und so weiter. Makes things get really ridiculous very quickly.
And they are.
This technology is looking to be on the market by the end of 2009/start of 2010.
So fucking what, I hear you say.
Well, here's a picture of the first ever handset based on this technology:
It was launched about a month and a half ago by Ericsson (no, not Sony Ericsson...).
Now it's not going on sale soon and I'm not going to give too many technical details (as the big security man with the stick may beat me over the head) but *ahem* according to the press releases, it can manage up to 150Mbps download (yes, there's talk of 300Mbps, but that's really a theoretical limit).
Now, I'm not going to either confirm or deny how fast exactly it can go, but all I'm going to say is that it had no bother working as a modem for a HD TV streaming demonstration at CeBIT a few weeks ago.
This shit is fast.
And it's working right now.
Of course, it won't be in your hands for a bit, but there's commitments to have these networks up and running with the earlier versions running at 100Mbps download and 50Mbps upload by 2010. That's over national distances. Can't tell you where (again, big man, stick, etc...), but compared to today's small-ish scale of deployment of HSUPA (highest speed 7.2Mbps), it's a massive revolution in this technology.
Anyway, here endeth the lesson, there wasn't really much of a point to this, other than to point out that within two years, the download speed of your mobile phone is probably going to go up by about 30-ish times if you have a techno-geek phone like the N95.
No point to all this speed in Ireland for years to come when today you get completely flahed by Irish networks for using god-damn basic GPRS crawling through the 'net' like teletext
That can only last so long.
I've worked out that it's possible for me to spend €3,000 in an hour on Pay As You Go over here.
This can't and won't last. Things are changing quickly. The flat rate tariffs for data are coming, there's ones for 3-10gb in the UK and Germany, they'll be in Ireland soon enough (may even be there now) and there's talk about the EU capping the costs like they have done with international calls.
Not to mention, on the speed of broadband roll out in Ireland, there's a lot of money to be made from this.
Okay the current high speed connection is called HSDPA or whatever right?
my question is, what will be the name of this connection be called, or are you not at liberty to say?
Or is it just going to be plain oul MIMO.
Man i feel like im at a seminar or something.
Ok, here's what going to happen:
HSDPA : High Speed Download Packet Access
That's what we have right now, the standard implementation is either 1.8Mbps or 3.6Mbps download and usually 384Kbps upload (the old "3G" speed)
What's next is HSUPA. The U bit stands for uplink.
This is already out in the UK, standards vary, but as a rule of thumb, for the moment, its 7.2Mbps download, 1.44Mbps upload.
Vodafone have been running this for about 4-5 months in the Uk, but there aren't that many handsets that support it yet.
Now things get a bit fuzzy.
The next generation is referred to as HSPA. Yup, we get rid of the U and the D and just call it High Speed Packet Access.
Anyway, this is between 14Mbps and 28Mbps-ish.
Upload figures are between 5.7 and about 11Mbps
As I said, terminology gets a bit wooly right about here. Even the 3G naming gets a bit weird, HSDPA is referred to as 3.5G, HSUPA is (or at least was) referred to as 3.75G and LTE was referred to as 3.9G, but as you'll see, there's a bit between 3.75 and 3.9G, so its very confusing.
This is coming to the UK by the end of this year, start of next year.
So.... next comes HSPAevo (christ, another bloody HSPA, yep, probably 3.876692G). The evo bit stands for really fast car (or evolution...).
Its also known as
Bloody nightmare, this naming stuff.
By this stage, the entire phone network will essentially be run end to end using IP.
This stuff maxes out at about 50Mbps, but will probably never get that far.
As far as I know, this is going to be deployed about mid 2009. All of these (HSUPA, HSPA, HSPAevo) are essentially upgrades to the standard 3G base station, from a network point of view and don't necessarily require a huge investment on the behalf of the operators. They're a pain in the arse for the mobile phone designers though, so if anythign holds this shit up, it'll be the handsets.
Finally, we get on to this LTE stuff. Some people call it 4G, I presonally think that's ridiculous, but then this naming and "G" stuff has gotten so ridiculous, I've given up complaining about it.
MIMO is the name used for the new technology. Its more than just a simple upgrade to the base station, it will probably requires a bit of new hardware but again, it can be done crappily with a mediocre upgrade.
This MIMO thing changes how the network works as instead of having one antenna on one phone talking to one antenna on the base station (as is how it works now), you'll have multiple antennas talking to multiple antennas.
This shit gets complicated.
So, in summary, this connection is called LTE and it won't be out till about 2010 (although DoCoMo in Japan, who call it Super3G, more flipping names, want to have it up and running by the end of next year).
Still though, without major delays, this technology should be available somewhere in the world within 2 years.
Which is a hell of a lot faster than Eircom's broadband roll out...
One thing that struck my mind.
If you have 4 antennas built into your phone, then it stands that this type of connection will need 4 times more bandwidth than 3G right?
For it work properly i mean, so that some fellas with in a mile of you do not hog your bandwidth.
Sounds promising in theory but in a place like ireland theyll probably cheap out on the number of base stations and could cause slower speeds than expected.
Unless the network works in a more efficent way than previous HSD/PA networks , enlighten me if im wrong here.
It works in a far more efficient way.
Hmm, I'm pondering whether or not to explain PSK or QAM here.
No, best not to. I had started, but I backed off. It's too much to go through on here. At least, not today. Maybe another day.
Anyway, in a very simplistic way of looking at it, if instead of just sending 1 0 1 0 as bits, you could instead send groups of bits, 101 001 010 111 and so on.
Well, what happens is by using some very clever and difficult electronics, you increase the number of bits in that group.
Thus, you increase the speed of transmission.
This means that 4 times more antennas will give 16 times more bandwith.
4 antennas isn't being proposed right now, only 2 are on the table at the moment, but the N95, for example, has 5 antennas for various bits n pieces (FM radio, WiFi/Bluetooth, 3G, 2G and GPS). It's the stuff behind, the electronics that makes sense of it all that's difficult. That takes a huge amount of processing power. Once that's sorted, you're most of the way there.
This isn't much of a problem for base stations as they're big and you can pump power into them but unfortunately, mobile phones have little power and aer very small and so you can't put a few Core Duos and a couple of cooling fans in them...
It's why I said it's more of a software upgrade for the base stations. It's not entirely, but they've (mostly) been kitted out to take it.
That makes sense to me anyhoo the binary side of it and sounds pretty intersting.
I wonder will this use of "groups" of bits will be used in other forms of communaction.
Is it a case of each antenna recieving its own bit?
Hence say 3 antennas, = 101.
Me thinkings its more advanced that that somehow though.
And are will the transmission frequency be higher? I.e a slightly shorter wavelength.
If its mostly a software trick probably not.
Ok, here's a picture that sort of illustrates it:
i may explain that properly yet, it's to do with complex numbers if you know anything about them...
If you have more antennas, you can have more dimensions and hence more possible bandwidth
Here's an example:
Say you've got a handset with two antennas, antenna A and antenna B
You're talking to a basestation with two seperate antennas, antenna 1 and antenna 2.
With a normal system, Antenna A talks to antenna 1, antenna B talks to antenna 2 and ne'er the twain shall meet.
Now with this MIMO (multiple input, multiple output, remember):
Antenna A can talk to 1 and 2 at the same time
Antenna B can talk to 1 and 2 at the same time.
The reverse also applies, 1 can talk to A and B and 2 can talk to A and B.
Now you've got 4 paths for data.
So if you have A, B and C in a phone and 1,2 and 3 in the base station, you have 9 possible data paths.
And so on. It's a square law.
So, between squeezing more 101 patterns into each transmission and making more paths for these transmissions, the data rates skyrocket without massive incrases in the infrastructure and without necessarily increasing the frequencies used.
Now how that last bit is done is, well, here's a ready made explanation from Wikipedia (I'm tired and it's almost home time..)
An analogy to the problem of multiple access is a room (channel) in which people wish to communicate with each other. To avoid confusion, people could take turns speaking (time division), speak at different pitches (frequency division), or speak in different directions (spatial division). In CDMA, they would speak different languages. People speaking the same language can understand each other, but not other people. Similarly, in radio CDMA, each group of users is given a shared code. Many codes occupy the same channel, but only users associated with a particular code can understand each other.