Tmobile to use 4x2 MIMO technology in LTE deployment

[joshnys8913]

T-mobile still has A LONG ways to go....

[milan03]

*looks at Sprints active 8T8R B41 deployment*

 

At least they didn't claim they are the absolute first...

 

 

 

 Anyhow it'll be a big boom to t-mobile subscribers where they get a signal and hopefully push other vendors to get out advanced mimo equipment faster and in larger numbers. Advancement and implementation of new technologies are always so good to see. 

 

Is there another wireless operator in the US that has a live 4x2 network today? To my knowledge there isn't.

 

I know Sprint's been talking about 8Tx/8Rx capable equipment lately and that's great, but afaik Sprint doesn't have commercially live Higher Order MIMO LTE Network.

 

Please correct me if I'm wrong.

[digiblur]

T-mobile still has A LONG ways to go....

 

Yes indeed, as well as Sprint (waiting to see B25 in many areas myself).

[WiWavelength]

Is there another wireless operator in the US that has a live 4x2 network today? To my knowledge there isn't.

 

I know Sprint's been talking about 8Tx/8Rx capable equipment lately and that's great, but afaik Sprint doesn't have commercially live Higher Order MIMO LTE Network.

 

Please correct me if I'm wrong.

 

"If a tree falls in a forest and no one is around to hear it, does it make a sound?"

 

If T-Mobile deploys higher order MIMO and no current UE can use it, does it matter?

 

AJ

[milan03]

"If a tree falls in a forest and no one is around to hear it, does it make a sound?"

 

If T-Mobile deploys higher order MIMO and no current UE can use it, does it matter?

 

AJ

AJ, all existing commercially available T-Mobile LTE devices are compatible with 4x2 MIMO LTE networks.

It's 4 transmit paths at the BTS, and the existing 2 receiving paths on the UE. Hence 4x2.

4x2 setup improves median down and up links, edge of cell performance and overall robustness and spectral efficiency of the network.

 

It's also a sidestep to 4x4 MIMO, so once UE manufacturers figure out how to muscle in 4 receive antennas without interference and poor battery life into that 4x4 capable device, T-Mobile's LTE users will instantly get the benefit of higher peak data rates of 4 spatial streams. That's 300Mbps peak download throughput, up from their existing 150Mbps in 20Mhz FDD LTE channels.

[WiWavelength]

AJ, all existing commercially available T-Mobile LTE devices are compatible with 4x2 MIMO LTE networks.

It's 4 transmit paths at the BTS, and the existing 2 receiving paths on the UE. Hence 4x2.

4x2 setup improves median down and up links, edge of cell performance and overall robustness and spectral efficiency of the network.

 

It's also a sidestep to 4x4 MIMO, so once UE manufacturers figure out how to muscle in 4 receive antennas without interference and poor battery life into that 4x4 capable device, T-Mobile's LTE users will instantly get the benefit of higher peak data rates of 4 spatial streams. That's 300Mbps peak download throughput, up from their existing 150Mbps in 20Mhz FDD LTE channels.

 

Hi Milan...

 

I get all of that.  But let us be honest about it.  With available devices, the current T-Mobile deployment is not higher order MIMO -- it is just spatial diversity.  Now, that is a good thing.  However, is it really different from what the former Clearwire wing of Sprint is doing?  We already have evidence of 8 Tx antenna infrastructure.  Do we really think that it is not using spatial diversity, too?

 

AJ 

[milan03]

Hi Milan...

 

I get all of that.  But let us be honest about it.  With available devices, the current T-Mobile deployment is not higher order MIMO -- it is just spatial diversity.  Now, that is a good thing.  However, is it really different from what the former Clearwire wing of Sprint is doing?  We already have evidence of 8 Tx antenna infrastructure.  Do we really think that it is not using spatial diversity, too?

 

AJ

 

AJ, this isn't spatial diversity only. Multiplexing on 2 spatial streams is still present in 4x2 setup. And with 4 streams at the tower top, there is a higher probability of spatial multiplexing altogether. So the probability of UE connecting to TM3 or TM4 is much greater in 4x2 setup.

[digiblur]

AJ, this isn't spatial diversity only. Multiplexing on 2 spatial streams is still present in 4x2 setup. And with 4 streams at the tower top, there is a higher probability of spatial multiplexing altogether. So the probability of UE connecting to TM3 or TM4 is much greater in 4x2 setup.

 

I always wondered why they had those weird looking antennas for those NSN setups, makes total sense now.  I see one on my daily commute right off the interstate, one of those sites that is still just Sprint legacy with no new equipment.

[milan03]

I always wondered why they had those weird looking antennas for those NSN setups, makes total sense now.  I see one on my daily commute right off the interstate, one of those sites that is still just Sprint legacy with no new equipment.

Some NSN sites have the thicker square looking 4-beam antennas, in other cases they would cluster two 2-beam antennas. Since NSN Flexi base stations are 4x4 MIMO capable, T-Mobile NSN had 4x4 capable antenna setup from the get go.

 

Now Ericsson markets are completely different story. Deploying 4x2 there would require new panels, upgrade to the BTS, etc... Don't think it'll happen this year.

[digiblur]

Some NSN sites have the thicker square looking 4-beam antennas, in other cases they would cluster two 2-beam antennas. Since NSN Flexi base stations are 4x4 MIMO capable, T-Mobile NSN had 4x4 capable antenna setup from the get go.

 

Now Ericsson markets are completely different story. Deploying 4x2 there would require new panels, upgrade to the BTS, etc... Don't think it'll happen this year.

 

Yeah, they have those really wide looking antennas.  Almost look like the 33 or 45 degree ones that Sprint has used.  They do have 3 bands in them as I can see the 3 remote downtilt controllers in them, makes sense now for the 4x4.  The site I see is an odd setup where they made a 4 sector site out of a 3 sector crows nest.  The fourth sector is hanging down below.  I've seen a few of them where they replaced the crows nest with a true four sector mount though.

[jamisonshaw125]

[reedacus25]

Hey Digiblur, 4x2 live in Nola, at least in some parts, if not all. Live from the Causeway to the Lakefront Airport, I'm sure CBD too but I very rarely venture that far down. Should hear back from someone in Elmwood/River Ridge to see if it's live there.

 

Next trip to Baton Rouge I'll see how Metaire and Kenner look.

[bucdenny]

Would Sprint benefit from 4x4 MIMO setup for B25 and B26? Would it increase the theoretical speed of 37mb?

[milan03]

Would Sprint benefit from 4x4 MIMO setup for B25 and B26? Would it increase the theoretical speed of 37mb?

Any network would benefit from 4x4 MIMO deployment, but UE antenna design with 4 (interference free) spatial streams is challenging, and so naturally handset OEMs still only produce 2x2 devices.

Luckily, there's been some progress, and we should hopefully see first handsets with 4x4 MIMO design by the end of this year and into 2015.

http://www.skycross.com/news-and-events/press-release/mwc-2014-4x4-mimo/

[bucdenny]

Any network would benefit from 4x4 MIMO deployment, but UE antenna design with 4 (interference free) spatial streams is challenging, and so naturally handset OEMs still only produce 2x2 devices.

Luckily, there's been some progress, and we should hopefully see first handsets with 4x4 MIMO design by the end of this year and into 2015.

http://www.skycross.com/news-and-events/press-release/mwc-2014-4x4-mimo/

Will existing Sprint NV 1.0 hardware support 4x4 MIMO when available via firmware upgrade?

[WiWavelength]

AJ, this isn't spatial diversity only. Multiplexing on 2 spatial streams is still present in 4x2 setup. And with 4 streams at the tower top, there is a higher probability of spatial multiplexing altogether. So the probability of UE connecting to TM3 or TM4 is much greater in 4x2 setup.

 

Again, I get that -- or I completely do not follow you.

 

My opinion does not count for much, but 4x2 MIMO is bastardized terminology.  MIMO should be reserved for discrete spatial streams, not spatial diversity.  So, 4x2 MIMO is really 2x2 MIMO repeated.  And in MIMO, 2x2 times two does not equal four discrete spatial streams.  In other words, 4x2 MIMO is 2x2 MIMO with spatial diversity, which is nothing new in the wireless industry.

 

Now, T-Mobile's 4x2 MIMO infrastructure may be superior to Sprint's 2x2 MIMO infrastructure.  No doubt.  However, T-Mobile is basically putting all of its eggs in one basket.  The band 4 LTE deployment is its end all, be all.  Sprint has more eggs in its basket.  And if we are going with the 4x2 MIMO parlance, then some Sprint Spark band 41 LTE deployment should rightly be deemed 8x2 MIMO.

 

AJ

[milan03]

Again, I get that -- or I completely do not follow you.

 

My opinion does not count for much, but 4x2 MIMO is bastardized terminology. MIMO should be reserved for discrete spatial streams, not spatial diversity. So, 4x2 MIMO is really 2x2 MIMO repeated. And in MIMO, 2x2 times two does not equal four discrete spatial streams. In other words, 4x2 MIMO is 2x2 MIMO with spatial diversity, which is nothing new in the wireless industry.

 

Now, T-Mobile's 4x2 MIMO infrastructure may be superior to Sprint's 2x2 MIMO infrastructure. No doubt. However, T-Mobile is basically putting all of its eggs in one basket. The band 4 LTE deployment is its end all, be all. Sprint has more eggs in its basket. And if we are going with the 4x2 MIMO parlance, then some Sprint Spark band 41 LTE deployment should rightly be deemed 8x2 MIMO.

 

AJ

AJ, I'm not comparing Sprint's infrastructure with T-Mobile's, that's never my intention. Sprint should be getting some fine equipment as we speak and once deployed, it should be high performance. Specially when paired with all that B41 spectrum.

Let's also not forget that T-Mobile is starting with 700a and PCS LTE deployment, so it's not AWS only. Many rural markets will have PCS and not AWS LTE as soon as this year. So that's getting interesting as well.

 

As for MIMO thing, again 4x2 MIMO is very much a MIMO technology, capable of sending 4 "discrete" spatial streams at the cell site. Now UE manufacturers still aren't producing a device with 4 receive antennas, but that doesn't in any way diminish the capabilities of the network.

 

In LTE terminology diversity utilization implies that spatial multiplexing isn't available (due to network limitations or unfavorable RF conditions), so the network scheduler assigns a single codeword for that particular UE. That's not what's happening here.

 

Also as of March on T-Mobile's 2x2 network, only about 35% of the time UE is utilizing MIMO, and that's mostly when in low mobility and with high SINR. With 4 streams at the tower top that percentage will be much higher. Sprint's B25 network has much less favorable MIMO numbers.

 

So the retainability of data streams and therefore MIMO utilization is significantly higher when 4x2 MIMO is active. That's also why the median data rates improve.

 

Sprints 8tx radio heads need core network to have higher order enabled, and as far as I know that hasn't happened yet.

[NYC126]

I thought Nokia Siemens will deploy sprint 8x2 mimo in the 2.5ghz band. I need clarification on this.Also I don't know much of the tech but I am seeing former clear wire sites with big radios behind the panels in New York city. I know it has to do with TDD LTE.

[bigsnake49]

I thought Nokia Siemens will deploy sprint 8x2 mimo in the 2.5ghz band. I need clarification on this.Also I don't know much of the tech but I am seeing former clear wire sites with big radios behind the panels in New York city. I know it has to do with TDD LTE.

 

You mean 8x8? All three vendors will deploy them for Sprint in the 2.5GHz band

[milan03]

You mean 8x8? All three vendors will deploy them for Sprint in the 2.5GHz band

Not necessarily. A single 8T8R radio per site could cover four 2x2 MIMO sectors, or two 4x4 MIMO sectors, etc... 

 

Typical cell site is three sector setup, and only if Sprint has 8T8R radio per sector, 8x2 MIMO is possible.

Considering how slow the equipment vendors are, chances are that Sprint may have 2x2 MIMO setup until they can deploy radio equipment in volume.

 

Also, as per John Saw's statement today, 8T8R equipment literally just exited the lab testing stage and went into the field testing stage, so it's safe to say that Sprint's TDD LTE Higher Order MIMO service isn't commercially live yet, as I've mentioned in my earlier posts.

[milan03]

Since now we know that Sprint is also going to be using 4x2 MIMO on their 8T8R sites, this video should be useful as it explains a recent research conducted by SRG on T-Mobile's live 4x2 MIMO network in North Dallas.