Give the man a prize! You hit the nail on the head. It's not about reaching your data cap faster -- it's that whatever you're downloading, you get it faster to 1. enjoy your device more 2. save battery life 3. improve network capacity so ALL users get better speeds (even those with older modems).
LTE-U small cells will be certified under Part 15 rules.
And in other places you tell us that LTE-U is a combination of a licensed channel with an unlicensed one. How can it be Part 15 if it is using cellular licensed channels that are covered by Part 22? The story keeps changing.
In the same way that a cellular phone is certified as being a Part 15 device (for Wi-Fi and Bluetooth) and certified as a cellular device for 2G/3G/4G. Multi-radio devices are very commonplace with smartphones being the most common. So an LTE-U small cell would get certified under the rules for cellular and the unlicensed radio transmissions would be certified under Part 15.
That Qualcomm -- YOU -- are creating a system that is deliberately incompatible with the existing WiFi users so that WiFi users are impacted to give your cell customers that wonderful "UX". That Qualcomm -- YOU -- make equipment that cell companies can install that use the private licensed frequencies for data, and now you want to jump into the unlicensed ones and act in a way that will degrade every other user's communications. I've already said that.
Not sure how you make the jump from Qualcomm wanting to enable the use of unlicensed spectrum by cellular operators, to Qualcomm wanting to "degrade every other user's communications". Degrade communications for customers of our own Wi-Fi products? Degrade communications to Wi-Fi access points that are built on Atheros chips? How would that benefit Qualcomm or the mobile operators? Forgetting the backlash and the bad PR that would result as a consequence of that (and incurring the wrath of the FCC), how would it make business sense for Qualcomm to harm its own products in the field?
Then you should have said that from the beginning, instead of give technobabble nonsense about how it is impossible for LTE-U to interfere with any existing WiFi and how Qualcomm doesn't have to care because the bands are so large.
I did say that. In my first response to your comment that mentioned "the fine article", I said the article is wrong and has done great harm by misinforming its readers. I wouldn't characterize my answers as "technobabble nonsense" - I was simply trying to provide you the technical facts. This is a technical discussion, and a simple exposition of the facts should make it quite obvious that there is no grand conspiracy here, and that both the design of the system and the empirical evidence show that LTE-U has no more of an impact on 5GHz Wi-Fi (to be very, very specific) than adding another Wi-Fi access point.
Existing WiFi services don't deliberately interfere with each other by using a different protocol that doesn't LBT or uses a shorter transmit gap.
Perhaps a basic understanding of radio communications could help in this discussion. Any kind of over the air transmission by one radiator robs other transmitters from capacity. Either by increasing the noise floor (thus reducing the SNR and reducing bitrate as dictated by Shannon theorem), or by reducing the amount of time that the other transmitters can use for their own transmission, in the case of LBT. So the Boingo access point in this example, if it is on the same channel as your personal hotspot, would have the effect of reducing the amount of time your hotspot can transmit over the air (because your hotspot, using LBT, has to back off and wait). Thus it reduces your throughput. Thus it is having an impact on your experience. QED.
if Qualcomm uses a protocol that doesn't follow the same fairness rules that everyone else does in a license-free public frequency band.
Apologies, but once again you are misinformed. Wi-Fi uses LBT, but LBT is not used by "everyone else" as you imply. Here's a list of radios that work in the unlicensed bands that do not use LBT:
Bluetooth and Bluetooth Low Energy
8021.15.4 (used for ZigBee and countless other proprietary protocols)
LTE-U is a proprietary system developed by the LTE-U group for use in North America.
LTE-U is built on the global 3GPP Release 10 specification for LTE. It is backed by the leading providers of LTE technology. How many backers does a technology need to go from "proprietary" to "standard"? 10? 100? 1000?
It takes advantage of the fact that the 5Ghz band is not regulated in the US.
False. The 5GHz band is regulated by FCC Part 15 rules, specifically subpart E (regulations for the Unlicensed National Information Infrastructure devices). Here's a link to the US regulations which govern LTE-U.
LTE-U does not implement the "Listen before talk" mechanism used in WiFi to allow stations to co-exists on the same frequency.
Neither does Bluetooth and Bluetooth Low Energy, 8021.15.4 (used for ZigBee and countless other proprietary protocols), cordless phones, AMIMON's 5GHz Wireless HDI "standard" (it's backed by only 13 companies - does that qualify it as a standard?), Nordic semiconductor's proprietary 2.4GHz transceivers, TI's proprietary 2.4GHz transceivers, Digi's proprietary 900MHz and 2.4GHz transceivers, every 2.4GHz radio used for RC cars/helicopters/drones, and countless others. Should I keep listing them?
Tests by companies like Intel show that CSAT can greatly impact packet latency in WiFi which impacts things like VoIP and streaming video
The former are given over to the belief that the unlicensed spectrum in the 2.4 and 5GHz band were set aside for their use while the latter having fat wallets can "buy their own damn spectrum"
I understand that this is the perception, but law, regulations, and precedent clearly show that this is not the case. I've given plenty of examples of companies other than cellular operators deploying for-profit networks in unlicensed spectrum - using Wi-Fi or otherwise.
It's also hard to not be skeptical that this would only be deployed at convention centers, malls, and the like and that it would be kept to the power levels allowed to unlicensed operators.
But there's a simple verification regime in place for this. It's called FCC certification. LTE-U equipment has to pass Part 15 rules in order to be certified. If they don't pass certification, they can't be bought/sold/deployed. Any equipment found violating these rules under which they are certified would incur the wrath of the FCC.
If you're at an airport and you try to use your phone as a mobile hotspot, the Boingo or iPass Wi-Fi routers in the airport will cause interference with your personal hotspot. Are Boingo and iPass supposed to take down their Wi-Fi access points too?
Inventors have to be compensated for the effort and investment they put into R&D. That's why patents exist. Otherwise there would be no motivation to invent and share those inventions.
If it is to have any range at all and be useful in any way, it will have to use more power than Part 15 allows. Otherwise it will just be yet another short-range WiFi service, which we already have. Why should it replace WiFi? If it isn't supposed to replace it, then don't put it in the same frequencies.
Any intentional RF radiator sold in the United States has to pass through FCC certification before it can be sold/bought/deployed. LTE-U small cells will be certified under Part 15 rules. That means an FCC-certified lab will test them for compliance with all aspects of Part 15 rules for the UNII bands, including power output. If they use more power than Part 15 allows, then they would not pass certification and cannot be sold/bought/deployed. But of course that won't be the case - because they are in fact designed to comply with Part 15 limitations on power output, so that they can pass FCC certification.
The goal is to improve the "UX", but only for cell users.
And a Wi-Fi ISP like Boingo that blankets an airport with Wi-Fi access point is only trying to improve the UX for their own customers and not anyone else. So what's your point?
From TFA:
The catch, of course, is that LTE-U's frequencies are already being used by every Wi-Fi access point in America,
And my access points are all 2.4GHz. I've found the 5GHz points to be few and far between. So much so that most of my devices don't have that built in. The statement can only mean it will use both 2.4G and 5G frequencies. Or should we ignore the obvious meaning of "every" and use instead the Qualcomm UX managed-experience definition?
I'm almost at a loss for words here. The article is wrong. This is not an "opinion", it's a question of facts. Verifiable facts. Here is the specification for the bands of operation for LTE-U straight out of the LTE-U spec document (available here for your perusing):
the LTE-U operation will be limited only to the following carrier frequencies for U-NII-1 and U-NII-3, respectively.
U-NII-1
o {f-0.2, f-0.1, f, f+0.1, f+0.2 | f = 5160, 5180, 5200, 5220, 5240} MHz
U-NII-3
o {f-0.2, f-0.1, f, f+0.1, f+0.2 | f = 5745, 5765, 5785, 5805, 5825} MHz
Is a direct quote from the actual LTE-U specification enough to correct the patently false information from the article? If this is not enough, then I don't know what would be.
This was not the case the CableLabs study which did indeed use a commercial implementation. Further, listen before talk in LTE-U is only now being a proposed as part of the standard. There are vendor specific implementations that already have this, but the standard does not and did not when the tests were done.
Are you referring to Slide 9 of the CableLabs presentation? What the chart there fails to show is Wi-Fi throughput in the presence of other Wi-Fi. That's the baseline. No one is saying that LTE-U will have zero impact on Wi-Fi throughput. But Wi-Fi impact on Wi-Fi is also not zero. What needs to be compared is LTE-U impact on Wi-Fi throughput versus Wi-Fi impact on Wi-Fi throughput. And the CableLabs presentation doesn't show that - so the data set is incomplete.
I'm glad you say you "realize this is completely legal by current regulations". Because it is.
I don't think mobile operators are asking for an exception to the rule. They're asking to abide by the same rules that already allow cable companies, hotels, airports, wireless ISPs etc. to benefit from this shared resource. That's all there is to it.
What does the frame structure have to do with complying with Part 15? From a physical layer perspective, the transmitter (the LTE-U small cell) abides by the maximum output power, maximum EIRP, and spectral density limits set in Part 15. Here are the rules.
I am not sure I understand your argument (not being facetious - just being honest).
In the cases you mentioned and in the case of LTE-U, commercial entities are using unlicensed spectrum to offer communication services. It's an open resource that by law is available for anyone to use as long as they abide by the rules on power output.
The case you mentioned about hotels blocking Wi-Fi is not applicable here. In that case the hotels were willfully interfering with hotspots. Nobody is advocating for willful interference here. Only the fair sharing of shared asset.
All those "commercial services" you trot out as excuses for allowing LTE-U to use the same bandwidth are different in one very significant way: they ALL obey the Part 15 rules for the use of the spectrum.
Huh? LTE-U will have to abide by the same Part 15 rules exactly. Did anyone claim anything different?
Those cable modem WiFi hotspots that are popping up don't have 100' towers and special high-power transmitters. Those Zigbee etc. systems for power monitoring etc. play well with others.
I think I've stated this repeatedly in other replies but good to repeat here again: LTE-U won't come from 100' tower with high-power transmitters. It's a small technology that is limited by the same Part 15 rules as other unlicensed users like Wi-Fi.
The other commercial difference is that those systems you talk about are either hidden completely from the user (power company, e.g.) or are sold explicitly as WiFi. A cellular company selling "LTE-U" services will not be telling the customers that they are now using public airspace with an existing body of both licensed and unlicensed users, and that the customer may find his phone unusable because of the four neighbors who have WiFi access points. The customer is going to expect the service he paid for, and he's going to be defrauded.
Why would the user find his phone unusable? The brilliant aspect of LTE-U is that it lets the network managed the user experience, which is a hallmark of cellular service. If the experience in unlicensed is degraded due to interference, the LTE-U small cell would simply provide more bandwidth to the user on the licensed LTE connection. That is something that is simply not possible with Wi-Fi alone.
Cellular services are covered by a different Part. Allowing their services to mix with the existing overloaded 2.4GHz users is a mistake and needs to be stopped.
Not going to use 2.4GHz. Going to only use UNII-1 and UNII-3 portions of the 5GHz unlicensed spectrum.
I realize that reading the fine article is a lost art here, but it was even mentioned in the summary, for God's sake. Here, from TFA:
Unsurprisingly, several outside experiments that pitted standard LTE technology or "simulated LTE-U" technology, in the case of one in-depth Google study, against Wi-Fi transmitters on the same frequencies found that LTE drastically reduced the throughput on the Wi-Fi connection.
What do you think "reduce the throughput" means? And what happens when there is more than one tower nearby? And when more and more phones start showing up there?
First, the Google study has been thoroughly discredited. As the "TFA" says, it used "simulated LTE-U" - literally signal generators that bear no resemblance to the actual LTE-U waveform. Google themselves admitted such on page 5 of their report. When those exact same tests were conducted using actual LTE-U equipment, the results were vastly different. You can check the results of the Google tests when real LTE-U equipment was used in this FCC filing.
Your statement about "more than one tower nearby" also betrays a lack of understanding of LTE-U, which is fine given that it's a new technology. LTE-U will not be transmitted from cell towers at all. It is a small cell technology, that has to abide by the exact same power output limits imposed by the FCC as Wi-Fi does in unlicensed spectrum.
That's out of hundreds of MHz of available bandwidth in the unlicensed bands.
The 2.4GHz WiFi band is about 100MHz wide (not "hundreds), and there are only FOUR non-overlapping 22MHz channels in that band. That means that anyone with a powerful signal on channel 1 will interfere with others who are using channels 1, 2, 3, 4 and to some extent, 5. Channel 6 overlaps channels 2, 3, 4, 5, 7, 8, 9 and 10. So, just FOUR LTE-U 22MHz signals could blanket the entire 2.4GHz WiFi band. With each smartphone using two channels...
Again, you misunderstand the technology. LTE-U is not targeted at 2.4GHz. At all. Period. It is targeted only at the 5GHz band. The FCC says there are 555MHz of bandwidth in the 5GHz band (check on fcc.gov here) I believe 555MHz qualifies as "hundreds of MHz". Not only so, LTE-U is only targeted at a fraction of that spectrum. That's because it's designed to only work in UNII-1 and UNII-3 portions of the UNII band. UNII-2 will be left entirely for Wi-Fi and other unlicensed users.
There would be no need to drown out anyone because (1) it's not technically possible given max LTE bandwidth limitation
You're claiming that it is impossible for a 20MHz LTE-U signal to interfere with a 22MHz WiFi signal. The entire point of TFA is that they have already been show to do so.
Again, remembering that LTE-U won't operate at all in 2.4GHz, that it leaves over 355MHz of UNII-2 spectrum completely untouched, and that at max it would occupy a total of 40MHz given the LTE standard, I stand by my earlier statement.
and (2) the plentiful spectrum that's available in the band.
100MHz is not "plentiful". One LTE-U signal can interfere with up to 9 WiFi channels. You ignore the fact that the LTE-U signal will be more powerful than a standard WiFi signal (because it has to cover a larger area) and is using a preferential sharing protocol that steals access from other users.
See earlier comments on the fact that LTE-U is not contending with Wi-Fi for 2.4GHz spectrum, and the the fact that LTE-U signals will be capped at the same maximum power as Wi-Fi and thus has similar range. This argument is based on incorrect information, but I ho
Similar range to Wi-Fi access points operating in the 5GHz band, since they have to abide by the same power output limitations set by the FCC.
What is the plan to deploy them?
Plan is to deploy them in high congestion areas like airports, malls, parks, busy street corners, etc.
How much of the unlicensed spectrum does it utilize, and is it in 2.4 or 5 ghz ranges.
A single LTE-U channel is only 20MHz wide. Up to two channels can be utilized, for up to 40MHz of bandwidth. All of this is only in the 5GHz band - LTE-U will not use the 2.4GHz band at all. There is 555 MHz of spectrum in the 5GHz band according to the FCC.
It is not intended to be deployed by users. It's intended to be deployed by mobile network operators in high congestions areas only. It is not a wide coverage technology, and it is not broadcast from cell towers.
Regardless of duty cycling, and/or other forms of "mitigation" I fail to see how occupying the same frequencies as our Wi-Fi routers can do anything other than steal capacity.
You are right - occupying the same frequency as an incumbent Wi-Fi router would reduce its throughput. No one is saying otherwise. But that would be true if you were adding another Wi-Fi access point, or any other radiator on the same frequency. The question becomes: for those operators that want to use unlicensed spectrum to increase the capacity of their networks, should they use Wi-Fi, or should they use LTE-U? Does LTE-U have any more of an impact on incumbent Wi-Fi than adding Wi-Fi instead?
The answer is: it has no more of an impact than adding another Wi-Fi access point. And in fact, because there is no mandated guidelines in the 802.11 spec for fair sharing of airtime between Wi-Fi access points, some are more aggressive than others. So the irony, and it has been shown in test after test, is that LTE-U's CSAT algorithm is more equitable than some Wi-Fi access points in the market today. So it is a better neighbor to Wi-Fi than some aggressive Wi-Fi access points. Case in point is this demo video of LTE-U in action: https://youtu.be/EalEd7fu_K0?t=20s
Further it is reported that LTE-U is more aggressive than Wi-Fi at grabbing open air time--shorter backoff period--meaning that where there is contention, it won't even play fair.
This is simply false. LTE-U is very deterministic about how much air time it grabs. First, an LTE-U small cell will scan to find unoccupied channels. If one is found, it'll use it. If no free channels are available, it'll pick the least occupied channel. It does that by listening for Wi-Fi beacons and other radiators on that channel. It'll keep listening to determine how many other Wi-Fi APs are already on that channel. Then it'll only take its proportion of the air time. This is akin to two people in a debate. It would be fair to give each side 50% of the time to speak their viewpoint. That's the approach that LTE-U takes.
You can check out the formula for calculating the air time on page 10 here: http://goo.gl/ZyYvQ4
In locations with already high-contention--like apartments, this sounds like a very unpalatable cocktail, enough to make Wi-Fi so slow as to be unusable. What are we missing?
What you're missing is that this is not targeted at apartment buildings or private homes. This is a technology that'll be surgically inserted into high congestion areas - airports, parks, malls, etc. - to add capacity where it's needed most. Because it operates in the 5GHz band with limited power output (because it abides by regulations for unlicensed spectrum), its range is similar to Wi-Fi. That is, if the mall across the street from your apartment has this deployed, the signal wouldn't be able to make it across the street to cause any issues with your own, or your neighbor's, Wi-Fi.
Slashdot Mirror
Give the man a prize! You hit the nail on the head. It's not about reaching your data cap faster -- it's that whatever you're downloading, you get it faster to 1. enjoy your device more 2. save battery life 3. improve network capacity so ALL users get better speeds (even those with older modems).
LOL - why would you expect my reply to be rude? I'm here to answer questions and be helpful.
LTE-U small cells will be certified under Part 15 rules.
And in other places you tell us that LTE-U is a combination of a licensed channel with an unlicensed one. How can it be Part 15 if it is using cellular licensed channels that are covered by Part 22? The story keeps changing.
In the same way that a cellular phone is certified as being a Part 15 device (for Wi-Fi and Bluetooth) and certified as a cellular device for 2G/3G/4G. Multi-radio devices are very commonplace with smartphones being the most common. So an LTE-U small cell would get certified under the rules for cellular and the unlicensed radio transmissions would be certified under Part 15.
That Qualcomm -- YOU -- are creating a system that is deliberately incompatible with the existing WiFi users so that WiFi users are impacted to give your cell customers that wonderful "UX". That Qualcomm -- YOU -- make equipment that cell companies can install that use the private licensed frequencies for data, and now you want to jump into the unlicensed ones and act in a way that will degrade every other user's communications. I've already said that.
Not sure how you make the jump from Qualcomm wanting to enable the use of unlicensed spectrum by cellular operators, to Qualcomm wanting to "degrade every other user's communications". Degrade communications for customers of our own Wi-Fi products? Degrade communications to Wi-Fi access points that are built on Atheros chips? How would that benefit Qualcomm or the mobile operators? Forgetting the backlash and the bad PR that would result as a consequence of that (and incurring the wrath of the FCC), how would it make business sense for Qualcomm to harm its own products in the field?
Then you should have said that from the beginning, instead of give technobabble nonsense about how it is impossible for LTE-U to interfere with any existing WiFi and how Qualcomm doesn't have to care because the bands are so large.
I did say that. In my first response to your comment that mentioned "the fine article", I said the article is wrong and has done great harm by misinforming its readers. I wouldn't characterize my answers as "technobabble nonsense" - I was simply trying to provide you the technical facts. This is a technical discussion, and a simple exposition of the facts should make it quite obvious that there is no grand conspiracy here, and that both the design of the system and the empirical evidence show that LTE-U has no more of an impact on 5GHz Wi-Fi (to be very, very specific) than adding another Wi-Fi access point.
Existing WiFi services don't deliberately interfere with each other by using a different protocol that doesn't LBT or uses a shorter transmit gap.
Perhaps a basic understanding of radio communications could help in this discussion. Any kind of over the air transmission by one radiator robs other transmitters from capacity. Either by increasing the noise floor (thus reducing the SNR and reducing bitrate as dictated by Shannon theorem), or by reducing the amount of time that the other transmitters can use for their own transmission, in the case of LBT. So the Boingo access point in this example, if it is on the same channel as your personal hotspot, would have the effect of reducing the amount of time your hotspot can transmit over the air (because your hotspot, using LBT, has to back off and wait). Thus it reduces your throughput. Thus it is having an impact on your experience. QED.
if Qualcomm uses a protocol that doesn't follow the same fairness rules that everyone else does in a license-free public frequency band.
Apologies, but once again you are misinformed. Wi-Fi uses LBT, but LBT is not used by "everyone else" as you imply. Here's a list of radios that work in the unlicensed bands that do not use LBT:
It won't. LTE-U shares the spectrum fairly with Wi-Fi. No more of an impact than adding a Wi-Fi access point.
You can see test results for yourself. Here are test results from a 3rd party (Signals Research Group) who tested LTE-U impact on Wi-Fi.
LTE-U is a proprietary system developed by the LTE-U group for use in North America.
LTE-U is built on the global 3GPP Release 10 specification for LTE. It is backed by the leading providers of LTE technology. How many backers does a technology need to go from "proprietary" to "standard"? 10? 100? 1000?
It takes advantage of the fact that the 5Ghz band is not regulated in the US.
False. The 5GHz band is regulated by FCC Part 15 rules, specifically subpart E (regulations for the Unlicensed National Information Infrastructure devices). Here's a link to the US regulations which govern LTE-U.
LTE-U does not implement the "Listen before talk" mechanism used in WiFi to allow stations to co-exists on the same frequency.
Neither does Bluetooth and Bluetooth Low Energy, 8021.15.4 (used for ZigBee and countless other proprietary protocols), cordless phones, AMIMON's 5GHz Wireless HDI "standard" (it's backed by only 13 companies - does that qualify it as a standard?), Nordic semiconductor's proprietary 2.4GHz transceivers, TI's proprietary 2.4GHz transceivers, Digi's proprietary 900MHz and 2.4GHz transceivers, every 2.4GHz radio used for RC cars/helicopters/drones, and countless others. Should I keep listing them?
Tests by companies like Intel show that CSAT can greatly impact packet latency in WiFi which impacts things like VoIP and streaming video
Please provide a link to the test results.
The former are given over to the belief that the unlicensed spectrum in the 2.4 and 5GHz band were set aside for their use while the latter having fat wallets can "buy their own damn spectrum"
I understand that this is the perception, but law, regulations, and precedent clearly show that this is not the case. I've given plenty of examples of companies other than cellular operators deploying for-profit networks in unlicensed spectrum - using Wi-Fi or otherwise.
It's also hard to not be skeptical that this would only be deployed at convention centers, malls, and the like and that it would be kept to the power levels allowed to unlicensed operators.
But there's a simple verification regime in place for this. It's called FCC certification. LTE-U equipment has to pass Part 15 rules in order to be certified. If they don't pass certification, they can't be bought/sold/deployed. Any equipment found violating these rules under which they are certified would incur the wrath of the FCC.
If you're at an airport and you try to use your phone as a mobile hotspot, the Boingo or iPass Wi-Fi routers in the airport will cause interference with your personal hotspot. Are Boingo and iPass supposed to take down their Wi-Fi access points too?
Inventors have to be compensated for the effort and investment they put into R&D. That's why patents exist. Otherwise there would be no motivation to invent and share those inventions.
Take a look at page 18 and page 25 in the LTE-U specification for LTE-U small cells: http://www.lteuforum.org/uploads/3/5/6/8/3568127/lte-u_forum_enb_minimum_requirements_for_lte_u_sdl_v1.0.pdf
If it is to have any range at all and be useful in any way, it will have to use more power than Part 15 allows. Otherwise it will just be yet another short-range WiFi service, which we already have. Why should it replace WiFi? If it isn't supposed to replace it, then don't put it in the same frequencies.
Any intentional RF radiator sold in the United States has to pass through FCC certification before it can be sold/bought/deployed. LTE-U small cells will be certified under Part 15 rules. That means an FCC-certified lab will test them for compliance with all aspects of Part 15 rules for the UNII bands, including power output. If they use more power than Part 15 allows, then they would not pass certification and cannot be sold/bought/deployed. But of course that won't be the case - because they are in fact designed to comply with Part 15 limitations on power output, so that they can pass FCC certification.
The goal is to improve the "UX", but only for cell users.
And a Wi-Fi ISP like Boingo that blankets an airport with Wi-Fi access point is only trying to improve the UX for their own customers and not anyone else. So what's your point?
From TFA:
The catch, of course, is that LTE-U's frequencies are already being used by every Wi-Fi access point in America,
And my access points are all 2.4GHz. I've found the 5GHz points to be few and far between. So much so that most of my devices don't have that built in. The statement can only mean it will use both 2.4G and 5G frequencies. Or should we ignore the obvious meaning of "every" and use instead the Qualcomm UX managed-experience definition?
I'm almost at a loss for words here. The article is wrong. This is not an "opinion", it's a question of facts. Verifiable facts. Here is the specification for the bands of operation for LTE-U straight out of the LTE-U spec document (available here for your perusing):
the LTE-U operation will be limited only to the following carrier frequencies for U-NII-1 and U-NII-3, respectively. U-NII-1 o {f-0.2, f-0.1, f, f+0.1, f+0.2 | f = 5160, 5180, 5200, 5220, 5240} MHz U-NII-3 o {f-0.2, f-0.1, f, f+0.1, f+0.2 | f = 5745, 5765, 5785, 5805, 5825} MHz
Is a direct quote from the actual LTE-U specification enough to correct the patently false information from the article? If this is not enough, then I don't know what would be.
This was not the case the CableLabs study which did indeed use a commercial implementation. Further, listen before talk in LTE-U is only now being a proposed as part of the standard. There are vendor specific implementations that already have this, but the standard does not and did not when the tests were done.
Are you referring to Slide 9 of the CableLabs presentation? What the chart there fails to show is Wi-Fi throughput in the presence of other Wi-Fi. That's the baseline. No one is saying that LTE-U will have zero impact on Wi-Fi throughput. But Wi-Fi impact on Wi-Fi is also not zero. What needs to be compared is LTE-U impact on Wi-Fi throughput versus Wi-Fi impact on Wi-Fi throughput. And the CableLabs presentation doesn't show that - so the data set is incomplete.
I'm glad you say you "realize this is completely legal by current regulations". Because it is.
I don't think mobile operators are asking for an exception to the rule. They're asking to abide by the same rules that already allow cable companies, hotels, airports, wireless ISPs etc. to benefit from this shared resource. That's all there is to it.
What does the frame structure have to do with complying with Part 15? From a physical layer perspective, the transmitter (the LTE-U small cell) abides by the maximum output power, maximum EIRP, and spectral density limits set in Part 15. Here are the rules.
Here you go...happy reading: http://goo.gl/ZyYvQ4
No different than paid Wi-Fi access points and wireless ISPs like Boingo and iPass.
It's a shared resource. I think people sometimes forget that.
They already have to deal with non-Wi-Fi interferers. For examples, Wireless HDI is a 5GHz technology that uses those same bands.
I am not sure I understand your argument (not being facetious - just being honest).
In the cases you mentioned and in the case of LTE-U, commercial entities are using unlicensed spectrum to offer communication services. It's an open resource that by law is available for anyone to use as long as they abide by the rules on power output.
The case you mentioned about hotels blocking Wi-Fi is not applicable here. In that case the hotels were willfully interfering with hotspots. Nobody is advocating for willful interference here. Only the fair sharing of shared asset.
I hope you've seen from my many replies now that I am here to answer questions.
All those "commercial services" you trot out as excuses for allowing LTE-U to use the same bandwidth are different in one very significant way: they ALL obey the Part 15 rules for the use of the spectrum.
Huh? LTE-U will have to abide by the same Part 15 rules exactly. Did anyone claim anything different?
Those cable modem WiFi hotspots that are popping up don't have 100' towers and special high-power transmitters. Those Zigbee etc. systems for power monitoring etc. play well with others.
I think I've stated this repeatedly in other replies but good to repeat here again: LTE-U won't come from 100' tower with high-power transmitters. It's a small technology that is limited by the same Part 15 rules as other unlicensed users like Wi-Fi.
The other commercial difference is that those systems you talk about are either hidden completely from the user (power company, e.g.) or are sold explicitly as WiFi. A cellular company selling "LTE-U" services will not be telling the customers that they are now using public airspace with an existing body of both licensed and unlicensed users, and that the customer may find his phone unusable because of the four neighbors who have WiFi access points. The customer is going to expect the service he paid for, and he's going to be defrauded.
Why would the user find his phone unusable? The brilliant aspect of LTE-U is that it lets the network managed the user experience, which is a hallmark of cellular service. If the experience in unlicensed is degraded due to interference, the LTE-U small cell would simply provide more bandwidth to the user on the licensed LTE connection. That is something that is simply not possible with Wi-Fi alone.
Cellular services are covered by a different Part. Allowing their services to mix with the existing overloaded 2.4GHz users is a mistake and needs to be stopped.
Not going to use 2.4GHz. Going to only use UNII-1 and UNII-3 portions of the 5GHz unlicensed spectrum.
I realize that reading the fine article is a lost art here, but it was even mentioned in the summary, for God's sake. Here, from TFA:
What do you think "reduce the throughput" means? And what happens when there is more than one tower nearby? And when more and more phones start showing up there?
First, the Google study has been thoroughly discredited. As the "TFA" says, it used "simulated LTE-U" - literally signal generators that bear no resemblance to the actual LTE-U waveform. Google themselves admitted such on page 5 of their report. When those exact same tests were conducted using actual LTE-U equipment, the results were vastly different. You can check the results of the Google tests when real LTE-U equipment was used in this FCC filing.
Your statement about "more than one tower nearby" also betrays a lack of understanding of LTE-U, which is fine given that it's a new technology. LTE-U will not be transmitted from cell towers at all. It is a small cell technology, that has to abide by the exact same power output limits imposed by the FCC as Wi-Fi does in unlicensed spectrum.
That's out of hundreds of MHz of available bandwidth in the unlicensed bands.
The 2.4GHz WiFi band is about 100MHz wide (not "hundreds), and there are only FOUR non-overlapping 22MHz channels in that band. That means that anyone with a powerful signal on channel 1 will interfere with others who are using channels 1, 2, 3, 4 and to some extent, 5. Channel 6 overlaps channels 2, 3, 4, 5, 7, 8, 9 and 10. So, just FOUR LTE-U 22MHz signals could blanket the entire 2.4GHz WiFi band. With each smartphone using two channels ...
Again, you misunderstand the technology. LTE-U is not targeted at 2.4GHz. At all. Period. It is targeted only at the 5GHz band. The FCC says there are 555MHz of bandwidth in the 5GHz band (check on fcc.gov here) I believe 555MHz qualifies as "hundreds of MHz". Not only so, LTE-U is only targeted at a fraction of that spectrum. That's because it's designed to only work in UNII-1 and UNII-3 portions of the UNII band. UNII-2 will be left entirely for Wi-Fi and other unlicensed users.
There would be no need to drown out anyone because (1) it's not technically possible given max LTE bandwidth limitation
You're claiming that it is impossible for a 20MHz LTE-U signal to interfere with a 22MHz WiFi signal. The entire point of TFA is that they have already been show to do so.
Again, remembering that LTE-U won't operate at all in 2.4GHz, that it leaves over 355MHz of UNII-2 spectrum completely untouched, and that at max it would occupy a total of 40MHz given the LTE standard, I stand by my earlier statement.
and (2) the plentiful spectrum that's available in the band.
100MHz is not "plentiful". One LTE-U signal can interfere with up to 9 WiFi channels. You ignore the fact that the LTE-U signal will be more powerful than a standard WiFi signal (because it has to cover a larger area) and is using a preferential sharing protocol that steals access from other users.
See earlier comments on the fact that LTE-U is not contending with Wi-Fi for 2.4GHz spectrum, and the the fact that LTE-U signals will be capped at the same maximum power as Wi-Fi and thus has similar range. This argument is based on incorrect information, but I ho
What are the ranges on the LTE-U base stations?
Similar range to Wi-Fi access points operating in the 5GHz band, since they have to abide by the same power output limitations set by the FCC.
What is the plan to deploy them?
Plan is to deploy them in high congestion areas like airports, malls, parks, busy street corners, etc.
How much of the unlicensed spectrum does it utilize, and is it in 2.4 or 5 ghz ranges.
A single LTE-U channel is only 20MHz wide. Up to two channels can be utilized, for up to 40MHz of bandwidth. All of this is only in the 5GHz band - LTE-U will not use the 2.4GHz band at all. There is 555 MHz of spectrum in the 5GHz band according to the FCC.
It is not intended to be deployed by users. It's intended to be deployed by mobile network operators in high congestions areas only. It is not a wide coverage technology, and it is not broadcast from cell towers.
Regardless of duty cycling, and/or other forms of "mitigation" I fail to see how occupying the same frequencies as our Wi-Fi routers can do anything other than steal capacity.
You are right - occupying the same frequency as an incumbent Wi-Fi router would reduce its throughput. No one is saying otherwise. But that would be true if you were adding another Wi-Fi access point, or any other radiator on the same frequency. The question becomes: for those operators that want to use unlicensed spectrum to increase the capacity of their networks, should they use Wi-Fi, or should they use LTE-U? Does LTE-U have any more of an impact on incumbent Wi-Fi than adding Wi-Fi instead?
The answer is: it has no more of an impact than adding another Wi-Fi access point. And in fact, because there is no mandated guidelines in the 802.11 spec for fair sharing of airtime between Wi-Fi access points, some are more aggressive than others. So the irony, and it has been shown in test after test, is that LTE-U's CSAT algorithm is more equitable than some Wi-Fi access points in the market today. So it is a better neighbor to Wi-Fi than some aggressive Wi-Fi access points. Case in point is this demo video of LTE-U in action: https://youtu.be/EalEd7fu_K0?t=20s
Further it is reported that LTE-U is more aggressive than Wi-Fi at grabbing open air time--shorter backoff period--meaning that where there is contention, it won't even play fair.
This is simply false. LTE-U is very deterministic about how much air time it grabs. First, an LTE-U small cell will scan to find unoccupied channels. If one is found, it'll use it. If no free channels are available, it'll pick the least occupied channel. It does that by listening for Wi-Fi beacons and other radiators on that channel. It'll keep listening to determine how many other Wi-Fi APs are already on that channel. Then it'll only take its proportion of the air time. This is akin to two people in a debate. It would be fair to give each side 50% of the time to speak their viewpoint. That's the approach that LTE-U takes.
You can check out the formula for calculating the air time on page 10 here: http://goo.gl/ZyYvQ4
In locations with already high-contention--like apartments, this sounds like a very unpalatable cocktail, enough to make Wi-Fi so slow as to be unusable. What are we missing?
What you're missing is that this is not targeted at apartment buildings or private homes. This is a technology that'll be surgically inserted into high congestion areas - airports, parks, malls, etc. - to add capacity where it's needed most. Because it operates in the 5GHz band with limited power output (because it abides by regulations for unlicensed spectrum), its range is similar to Wi-Fi. That is, if the mall across the street from your apartment has this deployed, the signal wouldn't be able to make it across the street to cause any issues with your own, or your neighbor's, Wi-Fi.