Amazon Is Testing Its Own Wireless Network

[kckid]

"Amazon.com Inc. (AMZN) has tested a new wireless network that would allow customers to connect its devices to the Internet, according to people with knowledge of the matter."

 

 

http://www.bloomberg.com/news/2013-08-23/amazon-is-said-to-have-tested-a-wireless-network.html

[richy]

It would fit with their ideology. They have the cash and could drive down their expenditures in the long run and allow them more freedom, but perhaps they're just goosing their current provider prior to the next contract negotiation?

[sbolen]

It's going to be really fun to be a part of the arms race that unfolds with Amazon, Apple, and Google in the content delivery space. If Amazon is looking to disrupt wireless, Google is looking to disrupt broadband, and Apple is looking to disrupt television, we, the consumer, will ultimately win.

[AJC1973]

i guess even amazon realilzes anything is better than att lol 

[bigsnake49]

Yeah, but somebody is going to have to build it for them. At WiFi frequencies, they will have to duplicate Clearwire's future site density. I'm suspicious!

[WiWavelength]

Yeah, but somebody is going to have to build it for them. At WiFi frequencies, they will have to duplicate Clearwire's future site density. I'm suspicious!

 

Well, the suspicion is that Amazon wants to offer a streaming/gaming appliance -- maybe even for free -- that will share Internet access with the outside world via potentially Globalstar's satellite spectrum converted to terrestrial spectrum.  Get a few of those Internet appliances per neighborhood, and fairly ubiquitous network access could be had that circumvents the incumbent wireless operators.

 

AJ

[bigsnake49]

Well, the suspicion is that Amazon wants to offer a streaming/gaming appliance -- maybe even for free -- that will share Internet access with the outside world via potentially Globalstar's satellite spectrum converted to terrestrial spectrum.  Get a few of those Internet appliances per neighborhood, and fairly ubiquitous network access could be had that circumvents the incumbent wireless operators.

 

AJ

 

Yes, but Globalstar TLPS spectrum is only 24MHz. I guess if you're going to use it as fixed broadband, you could turn up the power. Yes you could bypass the opearators that way, but it will cost you a pretty penny. Or you could rent capacity from Sprint on  EBS for cheaper, I think.

[WiWavelength]

Yes, but Globalstar TLPS spectrum is only 24MHz. I guess if you're going to use it as fixed broadband, you could turn up the power. Yes you could bypass the opearators that way, but it will cost you a pretty penny. Or you could rent capacity from Sprint on  EBS for cheaper, I think.

 

Renting BRS/EBS capacity would be cheaper?  Maybe, I do not know.  But the issue would be compatibility.  Good luck getting BRS/EBS added to most mobile devices.

 

Meanwhile, Globalstar's spectrum converted to terrestrial use potentially could be compatible with all existing Wi-Fi devices because it would unlock 2.4 GHz Wi-Fi channels 12-14 for licensed use.  Since other countries already use some/all of those channels above 11, current Wi-Fi devices may automatically be compatible with them.  That would be a huge advantage.

 

AJ

[bigsnake49]

Renting BRS/EBS capacity would be cheaper?  Maybe, I do not know.  But the issue would be compatibility.  Good luck getting BRS/EBS added to most mobile devices.

 

Meanwhile, Globalstar's spectrum converted to terrestrial use potentially could be compatible with all existing Wi-Fi devices because it would unlock 2.4 GHz Wi-Fi channels 12-14 for licensed use.  Since other countries already use some/all of those channels above 11, current Wi-Fi devices may automatically be compatible with them.  That would be a huge advantage.

 

AJ

 

If they want WiFi compatibility, then they're looking at low power operation...

[bigsnake49]

I'm looking forward to seeing what Dish is going to do with their spectrum as well as the end game of the Lightsquared saga. 

[WiWavelength]

If they want WiFi compatibility, then they're looking at low power operation...

 

Potentially.  But if Globalstar/Amazon were to be licensed the Wi-Fi spectrum, then power limits could be increased.  That probably would not extend to the uplink on existing Wi-Fi devices, though that might not matter.  It would depend upon how often existing Wi-Fi devices hit their uplink power limits under current circumstances.  Besides, the goal here would seem to be to crowdsource network access by getting millions of these Amazon appliances in homes and businesses across the country.  High power would not be needed because access point density would be high.

 

AJ

[bigsnake49]

Potentially.  But if Globalstar/Amazon were to be licensed the Wi-Fi spectrum, then power limits could be increased.  That probably would not extend to the uplink on existing Wi-Fi devices, though that might not matter.  It would depend upon how often existing Wi-Fi devices hit their uplink power limits under current circumstances.  Besides, the goal here would seem to be to crowdsource network access by getting millions of these Amazon appliances in homes and businesses across the country.  High power would not be needed because access point density would be high.

 

AJ

 

I don't understand what you're proposing here, AJ. Are you proposing using TLPS spectrum as backhaul to WiFi AP's?

[WiWavelength]

I don't understand what you're proposing here, AJ. Are you proposing using TLPS spectrum as backhaul to WiFi AP's?

 

No, Globalstar's satellite downlink spectrum is essentially 2.4 GHz Wi-Fi channels 12-14.  If the FCC were to allow that to be converted to TLPS, then it could be used for licensed Wi-Fi.  The Amazon appliances would offer licensed Wi-Fi to the public -- using consumers' broadband connections as backhaul.  The ringer is that current Wi-Fi devices could be compatible with the licensed channels.

 

http://www.fiercebroadbandwireless.com/story/globalstars-tlps-promises-private-licensed-wi-fi/2012-12-05

 

AJ

[bigsnake49]

No, Globalstar's satellite downlink spectrum is essentially 2.4 GHz Wi-Fi channels 12-14.  If the FCC were to allow that to be converted to TLPS, then it could be used for licensed Wi-Fi.  The Amazon appliances would offer licensed Wi-Fi to the public -- using consumers' broadband connections as backhaul.  The ringer is that current Wi-Fi devices could be compatible with the licensed channels.

 

http://www.fiercebroadbandwireless.com/story/globalstars-tlps-promises-private-licensed-wi-fi/2012-12-05

 

AJ

 

But then they could do that on the current unlicensed WiFi channels. If they want to have a nationwide WiFi network then can do that right now.

[WiWavelength]

But then they could do that on the current unlicensed WiFi channels. If they want to have a nationwide WiFi network then can do that right now.

 

Unlicensed 2.4 GHz Wi-Fi has become a logjam.  Interference is rampant, reducing range and speed.  If interference were basically eliminated by a move to licensed 2.4 GHz Wi-Fi, then range and speed would be greatly increased.  From reading a few FCC docs on the matter, that is the initial intent that I gather.

 

AJ

[bigsnake49]

Unlicensed 2.4 GHz Wi-Fi has become a logjam.  Interference is rampant, reducing range and speed.  If interference were basically eliminated by a move to licensed 2.4 GHz Wi-Fi, then range and speed would be greatly increased.  From reading a few FCC docs on the matter, that is the initial intent that I gather.

 

AJ

 

If you have too many APs, licensed spectrum will face the same problem as WiFi.

[WiWavelength]

If you have too many APs, licensed spectrum will face the same problem as WiFi.

 

I doubt it.  If the speculation were to come to fruition, the initial problem would be too few access points for decent coverage.  Then, as the number of access points grew, self interference could become a problem.  But Globalstar/Amazon would almost certainly use some self organizing network techniques to adjust power output per access point, even shut down unneeded TLPS on some access points in close proximity to others, since all access points would be under the control of one entity, unlike unlicensed Wi-Fi.  For example, see digiblur and his "boomer" Wi-Fi.

 

AJ

[bigsnake49]

I doubt it.  If the speculation were to come to fruition, the initial problem would be too few access points for decent coverage.  Then, as the number of access points grew, self interference could become a problem.  But Globalstar/Amazon would almost certainly use some self organizing network techniques to adjust power output per access point, even shut down unneeded TLPS on some access points in close proximity to others, since all access points would be under the control of one entity, unlike unlicensed Wi-Fi.  For example, see digiblur and his "boomer" Wi-Fi.

 

AJ

 

 

If you're going to be including all that logic, you might as well deploy a macro network. Like I said, I'm sceptical. I would like to see a lot more details.

[WiWavelength]

If you're going to be including all that logic, you might as well deploy a macro network. Like I said, I'm sceptical. I would like to see a lot more details.

 

TLPS does not seem to fit a macro network model.  Not enough power.  TLPS needs a site or access point every block.

 

AJ

[maximus1987/lou99]

TLPS does not seem to fit a macro network model. Not enough power. TLPS needs a site or access point every block.

 

AJ

Urban, suburban, rural?

[mozamcrew]

How about the FCC buy back the spectum currently licensed for chanels 12-14 so we can use that spectrum like everyone else?

[WiWavelength]

How about the FCC buy back the spectum currently licensed for chanels 12-14 so we can use that spectrum like everyone else?

 

What good would that accomplish?  Globalstar would want billions of dollars for the spectrum, and three additional unlicensed 2.4 GHz Wi-Fi channels, all of which overlap, would not solve any problems.

 

AJ

[mozamcrew]

What good would that accomplish?  Globalstar would want billions of dollars for the spectrum, and three additional unlicensed 2.4 GHz Wi-Fi channels, all of which overlap, would not solve any problems.

 

AJ

I thought channels 1,6, 11, and 14 were not overlapping using 20 Mhz channels (which is what 802.11g uses) N adds the option of doing 40 Mhz channels, but most of those are on the 5Ghz band since most multiband routers use 20Mhz channels on 2.4 for compatibility with older gear. So doing this would give you one more non overlapping channel  (And most gear supports this since in most countries those channels are part of the unlicensed spectrum)

[WiWavelength]

So doing this would give you one more non overlapping channel  (And most gear supports this since in most countries those channels are part of the unlicensed spectrum)

 

Like I said, one more non overlapping channel (channel 14) would not solve any unlicensed 2.4 GHz congestion problems.  It would be like putting a band aid on a bayonet wound.

 

AJ

[mozamcrew]

Like I said, one more non overlapping channel (channel 14) would not solve any unlicensed 2.4 GHz congestion problems.  It would be like putting a band aid on a bayonet wound.

 

AJ

 

I guess the question is how much benefit you get from going from 3 to 4 non overlapping channels. Surely there would be SOME benefit to having an additional channel. Maybe it's not the money (how much would that additional channel help with congestion at the cost of billions of dollars). But it would make our unlicenced band more like other countries, and it has the advantage of being usable by existing devices with a simple software update. Perhaps a spectrum swap is in order?