AllNet Labs/Fierce Wireless: "How much LTE spectrum do Verizon, AT&T, Sprint, and T-Mobile have - and where?"

[jamisonshaw125]

I think this successfully shows the difference in the amount of spectrum sprint has compared to....everyone else.

 

http://www.fiercewireless.com/special-reports/how-much-lte-spectrum-do-verizon-att-sprint-and-t-mobile-have-and-where

[JeremiahMcCurry]

Wow AT&T.

[CaryTheLabelGuy]

I think this successfully shows the difference in the amount of spectrum sprint has compared to....everyone else.

They failed to mention the additional 5x5Mhz B25 carrier that Sprint is deploying in certain markets. Although, I'm sure it was counted in the total.

[CaryTheLabelGuy]

Wow AT&T.

Oh yeah, AT&T is in a bad position spectrum-wise. They don't have a lot of room to grow and I'm sure they're really banking on the 600Mhz auction.

 

Sprint is the sleeping giant. Once this is all said and done, it'll be interesting to watch.

[JWMaloney]

They failed to mention the additional 5x5Mhz B25 carrier that Sprint is deploying in certain markets. Although, I'm sure it was counted in the total.

 

They also failed to map (but did mention) the additional 5x5MHz band 2 carrier that AT&T is deploying in certain markets.

 

Oh yeah, AT&T is in a bad position spectrum-wise. They don't have a lot of room to grow and I'm sure they're really banking on the 600Mhz auction.

 

Sprint is the sleeping giant. Once this is all said and done, it'll be interesting to watch.

 

They have another 15x15 of WCS almost nationally after their transaction with Sprint is completed. I believe they can use 10x10 of it without interfering with Sirius-XM.

[WiWavelength]

They failed to mention the additional 5x5Mhz B25 carrier that Sprint is deploying in certain markets. Although, I'm sure it was counted in the total.

 

I would not count on that.  AllNet Labs may have considered only Lower 700 MHz, Upper 700 MHz, PCS 1900 MHz G block, AWS 2100+1700 MHz, and BRS/EBS 2600 MHz as the de facto "4G" bands.  I may audit a few counties to confirm, but my initial visual inspection supports my theory.

 

AJ

[Nickel]

 

They also failed to map (but did mention) the additional 5x5MHz band 2 carrier that AT&T is deploying in certain markets.

 

 

They have another 15x15 of WCS almost nationally after their transaction with Sprint is completed. I believe they can use 10x10 of it without interfering with Sirius-XM.

Actually, while yes they do have a 10x10, the other 10 MHz came from two blocks of 5 MHz unpaired which they likely wouldn't have used anyway.

[CaryTheLabelGuy]

I would not count on that. AllNet Labs may have considered only Lower 700 MHz, Upper 700 MHz, PCS 1900 MHz G block, AWS 2100+1700 MHz, and BRS/EBS 2600 MHz as the de facto "4G" bands. I may audit a few counties to confirm, but my initial visual inspection supports my theory.

 

AJ

Interesting.

[NateC]

I think this successfully shows the difference in the amount of spectrum sprint has compared to....everyone else.

I would like to see a map like that excluding the 2.5 GHz spectrum.  Realistically we know that Sprint will only be deploying band 41 in urban areas since that's where it's needed.  In any case, the maps there make it hard to realistically compare AT&T/Verizon/Sprint in the non-urban areas since they're counting band 41 almost everywhere for Sprint.

[Destroyallcubes]

I would not count on that.  AllNet Labs may have considered only Lower 700 MHz, Upper 700 MHz, PCS 1900 MHz G block, AWS 2100+1700 MHz, and BRS/EBS 2600 MHz as the de facto "4G" bands.  I may audit a few counties to confirm, but my initial visual inspection supports my theory.

 

AJ

That map would change if they added in band 5 spectrum in the mix. If I'm not correct Verizon has some of the most spectrum that it holds in Dallas and through parts of east Texas, along I-20. Verizon Too. And Well it basically accounted for most of Tmobiles spectrum due to it mostly being AWS. And sprint because of spark

[WiWavelength]

Interesting.

 

Yeah, I appear to be correct.

 

The BRS/EBS spectrum holdings tend to swamp the other spectrum holdings, pushing the total to the maximum depicted on the AllNet Labs map.  So, for a case study, let us use an isolated rural county where Sprint has no native footprint and may not hold any BRS/EBS spectrum.  Union County, NM.

 

AllNet Labs shows that Sprint LTE downlink spectrum depth is in the 10-15 MHz range.  Well, in Union County, Sprint's spectrum holdings include at least the Dallas MTA PCS B block 30 MHz license, the Amarillo BEA PCS G block 10 MHz license, and the Amarillo BEA SMR X block 14 MHz rebanded license.

 

Thus, from an LTE downlink standpoint, that is 15 MHz + 5 MHz + 5 MHz = 25 MHz.  AllNet Labs does not reflect that.  Instead, it seems to count only the PCS G block and SMR X block spectrum.

 

AJ

[WiWavelength]

I would like to see a map like that excluding the 2.5 GHz spectrum.

 

I created such a map a decade ago.  It is still largely accurate.  So, just overlay nationwide PCS G block 10 MHz licenses and roughly nationwide SMR X block 14 MHz licenses.  There you have it...

 

 

http://people.ku.edu/~cinema/wireless/spcs_map.html

http://www.wirelesswavelength.com

 

AJ

[CaryTheLabelGuy]

Yeah, I appear to be correct.

 

The BRS/EBS spectrum holdings tend to swamp the other spectrum holdings, pushing the total to the maximum depicted on the AllNet Labs map. So, for a case study, let us use an isolated rural county where Sprint has no native footprint and may not hold any BRS/EBS spectrum. Union County, NM.

 

AllNet Labs shows that Sprint LTE downlink spectrum depth is in the 10-15 MHz range. Well, in Union County, Sprint's spectrum holdings include at least the Dallas MTA PCS B block 30 MHz license, the Amarillo BEA PCS G block 10 MHz license, and the Amarillo BEA SMR X block 14 MHz rebanded license.

 

Thus, from an LTE downlink standpoint, that is 15 MHz + 5 MHz + 5 MHz = 25 MHz. AllNet Labs does not reflect that. Instead, it seems to count only the PCS G block and SMR X block spectrum.

 

AJ

So in-other-words, they (AllNet) didn't really do their homework and the map can't be trusted as accurate, at least in Sprint's case.

 

Thanks for doing the legwork.

[WiWavelength]

So in-other-words, they (AllNet) didn't really do their homework and the map can't be trusted as accurate, at least in Sprint's case.

 

Thanks for doing the legwork.

 

Yes, AllNet Labs seems to be painting the 2014 LTE deployments with a broad brush.  That is nice for overview purposes, but it is not entirely accurate.  The map does not appear to take into account, for example, the AT&T band 2 LTE 1900 deployments in a few areas or the Sprint band 25 LTE 1900 second carrier in Chicago and coming soon to other markets.

 

This is why, when I create maps, I do not use automated data gathering methods.  I collect the data by hand and vet it against my own knowledge.  Of course, that takes hours on end, so it is also a reason why I do not make many maps any longer.

 

AJ

[JWMaloney]

So in-other-words, they (AllNet) didn't really do their homework and the map can't be trusted as accurate, at least in Sprint's case.

 

Thanks for doing the legwork.

 

I wouldn't say that. They didn't count PCS A-F for any carriers on their maps, and even if they had, the implication is still clear: Sprint is in the best spectrum position.

[WiWavelength]

...the implication is still clear: Sprint is in the best spectrum position.

 

Maybe not.  Go read the comments following the Fierce Wireless article.  Yeah, we have several obvious engineers and physicists stating that Sprint's BRS/EBS spectrum is "garbage," "can't penetrate its way out of a paper bag," and "will require a gazillion towers."  How fortunate we are to receive their expertise.

 

 

AJ

[JWMaloney]

Maybe not.  Go read the comments following the Fierce Wireless article.  Yeah, we have several obvious engineers and physicists stating that Sprint's BRS/EBS spectrum is "garbage," "can't penetrate its way out of a paper bag," and "will require a gazillion towers."  How fortunate we are to receive their expertise.

 

 

AJ

 

Go get 'em! Duty calls.

 

http://xkcd.com/386/

[Fraydog]

Maybe not. Go read the comments following the Fierce Wireless article. Yeah, we have several obvious engineers and physicists stating that Sprint's BRS/EBS spectrum is "garbage," "can't penetrate its way out of a paper bag," and "will require a gazillion towers." How fortunate we are to receive their expertise.

 

 

AJ

Spectrum turns into garbage north of 2170 MHz in their view. It's a magic barrier where spectrum simply becomes useless.

[cletus]

I wouldn't say that. They didn't count PCS A-F for any carriers on their maps, and even if they had, the implication is still clear: Sprint is in the best spectrum position.

But doesn't this also mean that Sprint has the most under-utilized spectrum and therefor needs the upcoming auctioned spectrum the least? Or are they making the argument that lower frequency = cheaper coverage from less towers and so the duopoly shouldn't get to bid? 

 

I mean... having a ton of spectrum doesn't help much when you can't get reliable speeds due to delays/backhauls/weather/prime contractor being shitty/etc

[WiWavelength]

Spectrum turns into garbage north of 2170 MHz in their view. It's a magic barrier where spectrum simply becomes useless.

 

Yes, Magentans are in AWE over AWS.

 

AJ

[JWMaloney]

But doesn't this also mean that Sprint has the most under-utilized spectrum and therefor needs the upcoming auctioned spectrum the least?

In my opinion, yes. Unless Sprint is planning to build out a rural footprint that rivals Verizon's, they don't need it -- they already have SMR spectrum.

 

Or are they making the argument that lower frequency = cheaper coverage from less towers and so the duopoly shouldn't get to bid?

I don't think the article was expressly trying to make any argument. I think it was just an objective piece on the big four's spectrum holdings to make people think about something other than speeds for once.

 

I mean... having a ton of spectrum doesn't help much when you can't get reliable speeds due to delays/backhauls/weather/prime contractor being shitty/etc

Ding-ding-ding!

[UserDemos]

AT&T has also refarmed Band 5 into LTE in places where there is no Band 17. Some are 5x5, some 10x10. Plenty of confirmed sightings. Those probably aren't accounted for either.

[WiWavelength]

AT&T has also refarmed Band 5 into LTE in places where there is no Band 17. Some are 5x5, some 10x10. Plenty of confirmed sightings. Those probably aren't accounted for either.

 

I doubt that AT&T has any 10 MHz FDD in band 5.  It would remove W-CDMA from that Cellular A/B block license.  Only where AT&T holds both Cellular A/B blocks would it have the wherewithal to deploy 10 MHz FDD.  And in those major markets, AT&T already holds Lower 700 MHz B/C block spectrum.

 

AJ

[nexgencpu]

Realistically we know that Sprint will only be deploying band 41 in urban areas since that's where it's needed.  In any case, the maps there make it hard to realistically compare AT&T/Verizon/Sprint in the non-urban areas since they're counting band 41 almost everywhere for Sprint.

This has been said repeatedly, but not true, in fact, a lot of users are finding 8t8r antenna's (Sprint B41) in the oddest places, urban and rural..

[NateC]

This has been said repeatedly, but not true, in fact, a lot of users are finding 8t8r antenna's (Sprint B41) in the oddest places, urban and rural..

 

Hmm, I'll accept that I might be wrong about that.  But, why?  From an engineering perspective that makes zero sense.  What's the purpose of rural B41 deployment?  The only reason I can think of is license protection.