No, the OP was right. The definition of 128-QAM is that it makes it possible to transmit 7 bits of data (128 = 2^7) per Hz of bandwidth. 64-QAM (2^6) is 6 bits per Hz.
You have to use QPSK (or 4-QAM) to transmit 2 bits of data per Hz.
If there were a modulation scheme that is twice as efficient as QAM that required the same SNR, everyone would be using it.
Frankly, the US should have taken the opportunity to create a nationwide LTE network infrastructure that all carriers could lease bandwidth/coverage from. There's no reason for verizon and at&t and others to build out the competing networks -- now we're stuck in the same boat as cable broadband where new providers have to bear the cost of building out areas even though they already have cable wired by another company.
Use a model where a private company builds out the network, and then leases it to any provider. More homogeneous technologies and frequencies used - intranationally and internationally - would be a boon as well.
The network backbone would sadly need to be a regulated monopoly, much like for other utilities, and the providers would be a competitive free for all.
This works for both cabling (fiber or coax to the home) and wireless services.
having lived in the US, corporations controlling telecommunications is the worst idea. it takes years to get new towers or upgrades to the network to new technology.
While one option might be for the government to build and manage the network, the point here really is to have a separate entity build and manage the physical network than the entities that provide service on that network. The purpose being to allow those that use the network to pool resources and function in a competitive environment where consumers could move between providers without changing hardware.
There could still be multiple types of networks each available to any provider that wanted to provide service on it. So you could still have a GSM network and a CDMA network... but if T-Mobile wanted they could offer a phone that roamed beween them both.
Using the popular "automotive analogy", what we have now is like have two sets of highways, one that is owned and built by Toyota paved with asphalt that only Toyotas can drive on and another is owned and built by GM paved with concrete that only GM's cars can drive on. This is incredibly inefficient and the consumer ends up paying for it because they are not as free as they should be to make a consumer coice to change. Consumers get to the end of the Toyota road only to see the GM road stretching further towards where they want to go, only they have to buy a new car to drive on it.
There is nothing technically impossible about building and selling phones that can roam between GSM & CDMA. It solely monopolistic practices at multiple levels designed to lock consumers in, those being restricted by contracts & restricted by single network hardware .
They run the network and any provider can lease access to that network and sell access to customers. All providers lease access using the same nationwide network and will compete on price and features offered instead of on coverage since spectrum is limited and finite.
It'll be interesting to see how LightSquared's network performs against Verizon and AT&T's LTE. It's good to see a company take this approach and solely build out the network as a dumb pipe and leave it up to the carriers to sell it to customers and compete.
I don't think that wireless is following the lead of Cable/DSL providers in charging more for higher speeds. Because wireless is so limited by spectrum they have pretty much almost all switched over to a metered usage plan. You might have also noticed that Verizon is doubling the GBs for LTE devices ONLY, making it effectively CHEAPER to user a higher speed service. I am guessing that their plan to make more money is to provide greater speed to encourage customer to use more and in turn choose larger buckets of data. This is probably best for the consumer because everyone gets to enjoy the best speeds, but only the heavy users pay more for it. Much more equitable and fair to all consumers.
"Though becoming the number one wireless provider in America has its appeal, the acquisition of spectrum is the driving force behind this deal."
Sorry Jason, but I have to disagree with you there. The driving force behind this deal is to remove a competitor from the market. It has already been proven via leaked legal documents that AT&T knows it has plenty of spectrum to build out its national LTE network.
According to fact, and arguments by AT&T as to why the merger should be allowed, Deutsche Telekom is not interested in investing further in T-Mobile or building out an LTE network to compete. There is industry speculation, and AT&T argument, that T-Mobile will fail if nothing is done. So if the purpose was to eliminate a competitor, simply waiting would be far cheaper. AT&T is just trying to assure itself of the spectrum and the customers before they get scattered to the winds.
Jason wrote: "That sounds logical, until you see how poorly US broadband speeds scale with their price; $200 for 105 Mbps just isn't balanced against $42 for 1.5 Mbps"
Are you saying we should pay $2,940 for 105 Mbps? That comes from $28/Mbps based on the low-speed pricing. (42/1.5*105)
Or are you saying we should only pay $2.87 for 1.5 Mbps? That's comes from $1.90/Mbps based on the high-speed pricing. (200/105*1.5)
What sort of scaling (balance) are you suggesting?
Is $200/105Mbps a good deal compared to $42/1.5Mbps? Perhaps, but you are comparing technologies separated by half a decade at least. Compare what $42 bought you in HD space 5 years ago to what it buys you today.
For a connection to just my phone that will be used sparingly, 1.5 Mbps is more than sufficient. 80-90% of the time when I can use it will be at work, home etc where a wifi connection will be superior.
The scaling shouldn't punish the low-end as much. Given the lack of competition in the broadband options, many people have a choice between dial-up ($20 for 42 kbps), DSL (~$40 for 0.5-1.5 Mbps) and Cable ($40 for 1.5 Mbps). It's rapidly becoming a necessity for reliable fast internet access in the home. Students unable to conduct their research on the internet from home are at a disadvantage. And with rich media resulting in webpages reaching the many megabytes on a regular basis it's not like dial-up is an acceptable option. So, $28/Mbps is the only realistic option for those unable to pay for more.
Verizon already charges more than other carriers for use of their network, so that's nothing new. Usually it's a $5-$10 extra per month versus competitors. Probably the most extreme is that:
Cricket offers a $55 monthly plan that includes unlimited talk, text, video, and 1GB data.
Verizon's best offer is $70 per month for 450 minutes, +20 cents per text, +25 cents per image/vid, and 2GB of data.
Unless you're on a family plan. I pay about $55/month for my portion, which includes about 500 minutes, unlimited text, and unlimited data (grandfathered in!). Plus we share 10 numbers (5 for individual plans) that are free to call so I rarely use more than 100 minutes/month. Unlimited texting is only $5/month to add to any plan so no one in there right mind would choose the per text charge.
Verizon's coverage and reliability is second to none. I gladly pay a bit more per month for quality service.
Verizon's network quality is why they charge the most out of any carrier. But for new customers like me that don't have plans to grandfather in, we get to pay $75 a month. What other option is there?
Instead of $75 a month, I am paying $100 for up to eight months of service, using prepaid. I'd love to have a smartphone but until I see a reasonable plan I'm not going to pay $1k a year for it when $200 will do.
Here in Sweden they have and use 2x20MHz LTE FDD in 2600-band and will be using 2x10MHz FDD for the 700-band and several companies have plenty of spectrum nationwide (although big, Sweden is sparsely populated and it wouldn't make sens in regionalising anything that way in the airways.) which still has no VoLTE up yet.
2x20MHz channels will yield some 40-80Mbits in good conditions, I really wouldn't expect more then 20-45Mbit on our future 800MHz LTE. Still superior to 3G, with lower latency but I won't expect it to built out that quickly when the infrastructure isn't upgraded yet and it needs to integrate with older generation systems and they also need larger backhaul which won't be available everywhere. That has to contend with 21Mbit or 42Mbit HSPA+ also, so it might just be about going from speeds about 10Mbit (±5) to 20Mbit (±10) and with relatively poor traffic quotas where many had virtually unlimited 3G subscription plans they won't let go of. Here in my small town there is just 64QAM HSPA+ Rel 7, larger town mostly have 64QAM HSPA+ Rel 7 (advertised as 32Mbit and not 4G mind you). If they don't anything about the traffic quotas those offers will just be too costly and they can't even promise (even though many would have) that much higher speeds. And a good 2600-band reception in a mayor city will still yield likely more then double the speed in your home town or residential neighborhood.
It kinda adds further confusion for the customers, in 3G (UMTS/WCDMA/HSPA/+) we know that we have 2x5MHz and what speeds there is kinda it still varies but it isn't because a frequency or spectrum divide. With LTE it will be different, and like if T-mobile would continue to operate and manged to get a reaction and win the C-block they would only get a 2x5MHz spectrum for LTE to use, that would literally be 20-25Mbit and they could easily match that with HSPA+ when it comes to speeds. Say they would acquire and launch in mayor metropolitan areas with 2x10 or 2x20 at higher bands then they would basically offer up to 100Mbit in some areas and up to 25 in others. Not that I think anyone in the us will go for the "up to 100Mbit LTE". But it would be a long way from our 100Mbit capable LTE in Sweden.
Other countries use standardized frequencies for 3G communication, and manage quite well despite facing the same bw use issues. The US is not unique in facing this problem.
The alternative to more spectrum is to build more smaller cells. The users/cell decreases, and thus, more bandwidth is available for each user.
" Unlike the macrocells, these small cells are distributed in an ad hoc manner, being placed in hotspots or coverage gaps. We've seen some of this in the deployment of picocells within AT&T and Apple stores in congested areas. "
If ATT were smart, - they'd build a picocell into the U-verse boxes they have on every block in U-verse cities. I see no evidence of this. - they'd make a deal with Apple, Google, and MS, that when phones have a WiFi connection they run the audio signal over WiFi --- just use the cell system for initial paging. Again no evidence of this.
I think the fundamental premise of this article: "ATT are good guys who want to provide customers with a better experience" is fundamentally false. ATT want T-Mobile to get rid of competition. If they were interested in a better customer experience, there are a variety of technical things they could do --- I've suggested two, but there are plenty more --- that they have zero interest in pursuing. ATT's goal is to perform at the bare minimum they can get away with without being sued --- they have ZERO interest in anything beyond that.
<insert carrier> is run by a pack of liars and cheats!
sarcasm aside, as a former AT&T customer, who got lied to on multiple occasions during my service period (and cancellation) I will admit that AT&T is terrible. Verizon is no saint, however, and considering their consistent attempts to hamstring their devices, from locking down GPS back in the mid 2000's, nickel-diming for every last feature, and today's locking of bootloaders on their smartphones.
Verizon seems to have the most consistent network... not necessarily the fastest 3g, but their 3G seems to have better coverage than AT&T in my travels. Verizon's LTE is just stupid fast though... can't wait for my next upgrade.
I hope that the fundamental premise of this article didn't come off as anything like you interpreted it. The fundamental premise, as I wrote it, was that "[m]ore hoses will be the answer every time, and in wireless that means more spectrum." AT&T is a corporation looking out for its corporate interests. As is Verizon. As is T-Mobile. In this case, Deutsche Telekom, T-Mobile's corporate parent, believes its best interest lies in selling T-Mobile. AT&T and Verizon believe their interests lie in obtaining spectrum.
Would it help AT&T to become the largest carrier and eliminate some of its competition? maybe, it certainly looks better in ads to be No. 1 rather than No. 2. But oneupmanship isn't actually all that valuable, and you can lose customers fast if your service is lacking, just ask Sprint. Verizon has about 107 million customers. AT&T has about 100 million customers. Sprint has about 53 million and T-Mobile has 33 million. T-Mobile isn't a competitor. They are fourth in a four man race, and their owner doesn't want to play anymore.
If AT&T hadn't made this deal, T-Mobile would still be for sale, and someone would still be looking to snap up their customers, and assets, for a bargain price. And if it was Verizon or Sprint, you'd still have the maintenance of the status quo and a competitor that will have to wrangle with the huge undertaking of bringing two companies together. More likely, a very complicated deal would have split the company's assets, with a small regional carrier (like Leap's Cricket) becoming the new fourth place.
But all that, is still an aside to the piece. Spectrum is hot again. We're not in a crunch, but for the first time in a while it is available and it is relatively cheap.
That's all fine and well, but I thought I read somewhere that cell phones cause cancer. So let's make the microwaves stronger and faster. Are we digging into the future... or digging our own graves?
AWS is about 1700MHz on the down link and 2100MHz on the uplink. T-Mobile has most of that space in the US, however, they have no plans to move to LTE. AT&T was planning to use that space to build out LTE, switching T-Mobile's 3G to AT&T's 3G bands (850MHz and 1900MHz).
Meanwhile, AT&T is building out its existing LTE network on its 700MHz spectrum.
In the 3rd picture (LTE_rates) I see column headings that I'm not clear on, say 2x2 on Downlink...what are those numbers? Since we're only looking at DL, it shouldn't have anything to do with UE transmit antennas...so perhaps NodeB x UE antennas or [UE]antennas x spatial streams...I'm unclear, as I said. Does LTE even allow/specify multiple spatial streams? I should know the answer...but I'm confident there are readers who do...a little help, please? Thanks!
I was wondering if the peak data rate for FDD and TDD based UE is significanly different? I mean, for example for FDD LTE Cat.3 device which can achieve 100/50 Mbps, what is the equivalent of peak data rate for TDD LTE Cat.3 device?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
38 Comments
Back to Article
TheCrackLing - Wednesday, December 7, 2011 - link
The article makes a mention of 128QAM as 2 more bits than 64QAM, however 128QAM only provides 7-bits, while 256QAM provides 8-bits.Sleepyhead5 - Wednesday, December 7, 2011 - link
Notice he said 2 bits per Hz, which is correct.For example, with a simple 1-bit modulation scheme it is possible to transmit 2 bits per Hz of bandwidth that you have at your disposal.
A5 - Wednesday, December 7, 2011 - link
No, the OP was right. The definition of 128-QAM is that it makes it possible to transmit 7 bits of data (128 = 2^7) per Hz of bandwidth. 64-QAM (2^6) is 6 bits per Hz.You have to use QPSK (or 4-QAM) to transmit 2 bits of data per Hz.
If there were a modulation scheme that is twice as efficient as QAM that required the same SNR, everyone would be using it.
KungFu_Toe - Wednesday, December 7, 2011 - link
Is there any info on the spectrum that AT&T offered to T-Mobile if the merger fell through?prophet001 - Wednesday, December 7, 2011 - link
"Verizon is also throwing in $100 million to help build out Cricket's LTE network."That's so nice of them to invest in their competitors....
RamarC - Wednesday, December 7, 2011 - link
Frankly, the US should have taken the opportunity to create a nationwide LTE network infrastructure that all carriers could lease bandwidth/coverage from. There's no reason for verizon and at&t and others to build out the competing networks -- now we're stuck in the same boat as cable broadband where new providers have to bear the cost of building out areas even though they already have cable wired by another company.alent1234 - Wednesday, December 7, 2011 - link
having worked for the US government, this is the worst idea. it would take years to put up a new tower or upgrade the network to a new technologyprophet001 - Wednesday, December 7, 2011 - link
thisZoomer - Wednesday, December 7, 2011 - link
Use a model where a private company builds out the network, and then leases it to any provider. More homogeneous technologies and frequencies used - intranationally and internationally - would be a boon as well.The network backbone would sadly need to be a regulated monopoly, much like for other utilities, and the providers would be a competitive free for all.
This works for both cabling (fiber or coax to the home) and wireless services.
cosmotic - Thursday, December 8, 2011 - link
having lived in the US, corporations controlling telecommunications is the worst idea. it takes years to get new towers or upgrades to the network to new technology.jb510 - Thursday, December 8, 2011 - link
thisap90033 - Thursday, December 8, 2011 - link
I agree if the US government were involved it would cost 13 TRILLION Dollars and we would be talking about dial up for "high speed" lol....jb510 - Thursday, December 8, 2011 - link
I think you're missing the point.While one option might be for the government to build and manage the network, the point here really is to have a separate entity build and manage the physical network than the entities that provide service on that network. The purpose being to allow those that use the network to pool resources and function in a competitive environment where consumers could move between providers without changing hardware.
There could still be multiple types of networks each available to any provider that wanted to provide service on it. So you could still have a GSM network and a CDMA network... but if T-Mobile wanted they could offer a phone that roamed beween them both.
Using the popular "automotive analogy", what we have now is like have two sets of highways, one that is owned and built by Toyota paved with asphalt that only Toyotas can drive on and another is owned and built by GM paved with concrete that only GM's cars can drive on. This is incredibly inefficient and the consumer ends up paying for it because they are not as free as they should be to make a consumer coice to change. Consumers get to the end of the Toyota road only to see the GM road stretching further towards where they want to go, only they have to buy a new car to drive on it.
There is nothing technically impossible about building and selling phones that can roam between GSM & CDMA. It solely monopolistic practices at multiple levels designed to lock consumers in, those being restricted by contracts & restricted by single network hardware .
defaultPlayer - Tuesday, December 13, 2011 - link
A private company is building out a national wholesale LTE network - LightSquaredhttp://www.lightsquared.com/
They run the network and any provider can lease access to that network and sell access to customers. All providers lease access using the same nationwide network and will compete on price and features offered instead of on coverage since spectrum is limited and finite.
It'll be interesting to see how LightSquared's network performs against Verizon and AT&T's LTE. It's good to see a company take this approach and solely build out the network as a dumb pipe and leave it up to the carriers to sell it to customers and compete.
mcnabney - Wednesday, December 7, 2011 - link
I don't think that wireless is following the lead of Cable/DSL providers in charging more for higher speeds. Because wireless is so limited by spectrum they have pretty much almost all switched over to a metered usage plan. You might have also noticed that Verizon is doubling the GBs for LTE devices ONLY, making it effectively CHEAPER to user a higher speed service. I am guessing that their plan to make more money is to provide greater speed to encourage customer to use more and in turn choose larger buckets of data. This is probably best for the consumer because everyone gets to enjoy the best speeds, but only the heavy users pay more for it. Much more equitable and fair to all consumers.DigitalFreak - Wednesday, December 7, 2011 - link
"Though becoming the number one wireless provider in America has its appeal, the acquisition of spectrum is the driving force behind this deal."Sorry Jason, but I have to disagree with you there. The driving force behind this deal is to remove a competitor from the market. It has already been proven via leaked legal documents that AT&T knows it has plenty of spectrum to build out its national LTE network.
CrapONez - Thursday, December 8, 2011 - link
According to fact, and arguments by AT&T as to why the merger should be allowed, Deutsche Telekom is not interested in investing further in T-Mobile or building out an LTE network to compete. There is industry speculation, and AT&T argument, that T-Mobile will fail if nothing is done. So if the purpose was to eliminate a competitor, simply waiting would be far cheaper. AT&T is just trying to assure itself of the spectrum and the customers before they get scattered to the winds.justaviking - Wednesday, December 7, 2011 - link
Jason wrote: "That sounds logical, until you see how poorly US broadband speeds scale with their price; $200 for 105 Mbps just isn't balanced against $42 for 1.5 Mbps"Are you saying we should pay $2,940 for 105 Mbps?
That comes from $28/Mbps based on the low-speed pricing. (42/1.5*105)
Or are you saying we should only pay $2.87 for 1.5 Mbps?
That's comes from $1.90/Mbps based on the high-speed pricing. (200/105*1.5)
What sort of scaling (balance) are you suggesting?
joshv - Wednesday, December 7, 2011 - link
Is $200/105Mbps a good deal compared to $42/1.5Mbps? Perhaps, but you are comparing technologies separated by half a decade at least. Compare what $42 bought you in HD space 5 years ago to what it buys you today.Zoomer - Wednesday, December 7, 2011 - link
For a connection to just my phone that will be used sparingly, 1.5 Mbps is more than sufficient. 80-90% of the time when I can use it will be at work, home etc where a wifi connection will be superior.JasonInofuentes - Wednesday, December 14, 2011 - link
The scaling shouldn't punish the low-end as much. Given the lack of competition in the broadband options, many people have a choice between dial-up ($20 for 42 kbps), DSL (~$40 for 0.5-1.5 Mbps) and Cable ($40 for 1.5 Mbps). It's rapidly becoming a necessity for reliable fast internet access in the home. Students unable to conduct their research on the internet from home are at a disadvantage. And with rich media resulting in webpages reaching the many megabytes on a regular basis it's not like dial-up is an acceptable option. So, $28/Mbps is the only realistic option for those unable to pay for more.Kougar - Wednesday, December 7, 2011 - link
Verizon already charges more than other carriers for use of their network, so that's nothing new. Usually it's a $5-$10 extra per month versus competitors. Probably the most extreme is that:Cricket offers a $55 monthly plan that includes unlimited talk, text, video, and 1GB data.
Verizon's best offer is $70 per month for 450 minutes, +20 cents per text, +25 cents per image/vid, and 2GB of data.
dcollins - Wednesday, December 7, 2011 - link
Unless you're on a family plan. I pay about $55/month for my portion, which includes about 500 minutes, unlimited text, and unlimited data (grandfathered in!). Plus we share 10 numbers (5 for individual plans) that are free to call so I rarely use more than 100 minutes/month. Unlimited texting is only $5/month to add to any plan so no one in there right mind would choose the per text charge.Verizon's coverage and reliability is second to none. I gladly pay a bit more per month for quality service.
Kougar - Sunday, December 11, 2011 - link
Verizon's network quality is why they charge the most out of any carrier. But for new customers like me that don't have plans to grandfather in, we get to pay $75 a month. What other option is there?Instead of $75 a month, I am paying $100 for up to eight months of service, using prepaid. I'd love to have a smartphone but until I see a reasonable plan I'm not going to pay $1k a year for it when $200 will do.
casteve - Wednesday, December 7, 2011 - link
Second paragraph under the QAM constellation, last sentence:"If you are lucky, your phone will have (if supported) a MIMO connection with low SNR..."
I think you meant to say high SNR.
Uritziel - Wednesday, December 7, 2011 - link
I guess I'm unlucky, as my phone does not support any connection with low SNR... ;)joshv - Wednesday, December 7, 2011 - link
"increasingly points on the plot"Increasingly what? Small? The adjective appears to be missing here.
Penti - Wednesday, December 7, 2011 - link
Here in Sweden they have and use 2x20MHz LTE FDD in 2600-band and will be using 2x10MHz FDD for the 700-band and several companies have plenty of spectrum nationwide (although big, Sweden is sparsely populated and it wouldn't make sens in regionalising anything that way in the airways.) which still has no VoLTE up yet.2x20MHz channels will yield some 40-80Mbits in good conditions, I really wouldn't expect more then 20-45Mbit on our future 800MHz LTE. Still superior to 3G, with lower latency but I won't expect it to built out that quickly when the infrastructure isn't upgraded yet and it needs to integrate with older generation systems and they also need larger backhaul which won't be available everywhere. That has to contend with 21Mbit or 42Mbit HSPA+ also, so it might just be about going from speeds about 10Mbit (±5) to 20Mbit (±10) and with relatively poor traffic quotas where many had virtually unlimited 3G subscription plans they won't let go of. Here in my small town there is just 64QAM HSPA+ Rel 7, larger town mostly have 64QAM HSPA+ Rel 7 (advertised as 32Mbit and not 4G mind you). If they don't anything about the traffic quotas those offers will just be too costly and they can't even promise (even though many would have) that much higher speeds. And a good 2600-band reception in a mayor city will still yield likely more then double the speed in your home town or residential neighborhood.
It kinda adds further confusion for the customers, in 3G (UMTS/WCDMA/HSPA/+) we know that we have 2x5MHz and what speeds there is kinda it still varies but it isn't because a frequency or spectrum divide. With LTE it will be different, and like if T-mobile would continue to operate and manged to get a reaction and win the C-block they would only get a 2x5MHz spectrum for LTE to use, that would literally be 20-25Mbit and they could easily match that with HSPA+ when it comes to speeds. Say they would acquire and launch in mayor metropolitan areas with 2x10 or 2x20 at higher bands then they would basically offer up to 100Mbit in some areas and up to 25 in others. Not that I think anyone in the us will go for the "up to 100Mbit LTE". But it would be a long way from our 100Mbit capable LTE in Sweden.
ravisurdhar - Wednesday, December 7, 2011 - link
Not to nitpick, but "where customers might not get any single from higher frequency towers." - should be "signal". ;)Thanks for the great articles, as usual.
Zoomer - Wednesday, December 7, 2011 - link
Other countries use standardized frequencies for 3G communication, and manage quite well despite facing the same bw use issues. The US is not unique in facing this problem.The alternative to more spectrum is to build more smaller cells. The users/cell decreases, and thus, more bandwidth is available for each user.
name99 - Thursday, December 8, 2011 - link
"Unlike the macrocells, these small cells are distributed in an ad hoc manner, being placed in hotspots or coverage gaps. We've seen some of this in the deployment of picocells within AT&T and Apple stores in congested areas.
"
If ATT were smart,
- they'd build a picocell into the U-verse boxes they have on every block in U-verse cities. I see no evidence of this.
- they'd make a deal with Apple, Google, and MS, that when phones have a WiFi connection they run the audio signal over WiFi --- just use the cell system for initial paging. Again no evidence of this.
I think the fundamental premise of this article: "ATT are good guys who want to provide customers with a better experience" is fundamentally false. ATT want T-Mobile to get rid of competition. If they were interested in a better customer experience, there are a variety of technical things they could do --- I've suggested two, but there are plenty more --- that they have zero interest in pursuing. ATT's goal is to perform at the bare minimum they can get away with without being sued --- they have ZERO interest in anything beyond that.
c4v3man - Thursday, December 8, 2011 - link
<insert carrier> is run by a pack of liars and cheats!sarcasm aside, as a former AT&T customer, who got lied to on multiple occasions during my service period (and cancellation) I will admit that AT&T is terrible. Verizon is no saint, however, and considering their consistent attempts to hamstring their devices, from locking down GPS back in the mid 2000's, nickel-diming for every last feature, and today's locking of bootloaders on their smartphones.
Verizon seems to have the most consistent network... not necessarily the fastest 3g, but their 3G seems to have better coverage than AT&T in my travels. Verizon's LTE is just stupid fast though... can't wait for my next upgrade.
JasonInofuentes - Saturday, December 10, 2011 - link
I hope that the fundamental premise of this article didn't come off as anything like you interpreted it. The fundamental premise, as I wrote it, was that "[m]ore hoses will be the answer every time, and in wireless that means more spectrum." AT&T is a corporation looking out for its corporate interests. As is Verizon. As is T-Mobile. In this case, Deutsche Telekom, T-Mobile's corporate parent, believes its best interest lies in selling T-Mobile. AT&T and Verizon believe their interests lie in obtaining spectrum.Would it help AT&T to become the largest carrier and eliminate some of its competition? maybe, it certainly looks better in ads to be No. 1 rather than No. 2. But oneupmanship isn't actually all that valuable, and you can lose customers fast if your service is lacking, just ask Sprint. Verizon has about 107 million customers. AT&T has about 100 million customers. Sprint has about 53 million and T-Mobile has 33 million. T-Mobile isn't a competitor. They are fourth in a four man race, and their owner doesn't want to play anymore.
If AT&T hadn't made this deal, T-Mobile would still be for sale, and someone would still be looking to snap up their customers, and assets, for a bargain price. And if it was Verizon or Sprint, you'd still have the maintenance of the status quo and a competitor that will have to wrangle with the huge undertaking of bringing two companies together. More likely, a very complicated deal would have split the company's assets, with a small regional carrier (like Leap's Cricket) becoming the new fourth place.
But all that, is still an aside to the piece. Spectrum is hot again. We're not in a crunch, but for the first time in a while it is available and it is relatively cheap.
Thanks for the comments, keep'em coming.
Jason
j03h4gLund - Monday, December 12, 2011 - link
That's all fine and well, but I thought I read somewhere that cell phones cause cancer. So let's make the microwaves stronger and faster. Are we digging into the future... or digging our own graves?Proph3T08 - Tuesday, December 13, 2011 - link
What are the major differences between these bands? If the US has plenty of AWS why are wireless carriers not focusing on using more of these bands?KPOM - Thursday, December 15, 2011 - link
AWS is about 1700MHz on the down link and 2100MHz on the uplink. T-Mobile has most of that space in the US, however, they have no plans to move to LTE. AT&T was planning to use that space to build out LTE, switching T-Mobile's 3G to AT&T's 3G bands (850MHz and 1900MHz).Meanwhile, AT&T is building out its existing LTE network on its 700MHz spectrum.
ChanceSF - Thursday, February 23, 2012 - link
In the 3rd picture (LTE_rates) I see column headings that I'm not clear on, say 2x2 on Downlink...what are those numbers? Since we're only looking at DL, it shouldn't have anything to do with UE transmit antennas...so perhaps NodeB x UE antennas or [UE]antennas x spatial streams...I'm unclear, as I said. Does LTE even allow/specify multiple spatial streams? I should know the answer...but I'm confident there are readers who do...a little help, please? Thanks!anat17 - Tuesday, March 5, 2013 - link
I was wondering if the peak data rate for FDD and TDD based UE is significanly different?I mean, for example for FDD LTE Cat.3 device which can achieve 100/50 Mbps, what is the equivalent of peak data rate for TDD LTE Cat.3 device?