5 ms·

Extending WiFi to one mile, thanks to empty TV channels

43 points by abraham 16y ago

dasil003 16y ago
And what does the signal look like if you tune a TV to that channel? Is it possible to fabricate a signal that shows up as a recognizable image?
- StavrosK 16y ago
  From what I gather, it's already very recognizable as moving specks of salt and pepper.
- wiredfool 16y ago
  Presumably, it looks like the sky over Chiba city. Maybe with a bit of the sprawl ghosting in it.
derobert 16y ago
"The longest link he could make with existing point-to-point WiFi connections was 400-500 feet; with the new Super WiFi gear in the TV band, he can reach a mile—and it's not a point-to-point signal." Something doesn't make sense here. This is the furthest they could get with point-to-point line of sight links, on towers. When I worked at a WISP, we routinely made longer PtP links—heck, even longer point-to-multipoint links. A mile should be easily doable with a clear line of sight. Using http://www.afar.net/rf-link-budget-calculator/ http://www.afar.net/rf-link-budget-calculator/ shows that a five-mile link should be no problem with 24dBi grid dishes (which are more or less the size of DirecTV, DISH, etc. dishes).
- wmf 16y ago
  "signals don't easily penetrate the tree branches and leaves which surrounded Aguirre's home" The whole point is non line of sight.
  - VladRussian 16y ago
    the phased array WiFi switches have interesting feature - adaptively found best beam direction to your computer may actually be not a straight line from the switch to you, instead it may happen to be a reflected beam.
    - wmf 16y ago
      Other than Vivato (who went broke), is there any shipping outdoor phased-array WISP equipment?
thaumaturgy 16y ago
Huh. Anybody have any idea what the FCC restrictions are on stuff like this?
- endian 16y ago
  "Guerra is part of a Rice team [...whose...] goal is basic research on providing broadband in the TV "white spaces" — empty channels that the government recently cleared for use by unlicensed Internet providers"
  - thaumaturgy 16y ago
    Yeah, I was hoping to find a little bit more in-depth information though. I'm finding reports of a handful of companies (including Google) being selected to be database "administrators" for this, but I can't get clarity on whether such a database is supposed to apply to equipment manufacturers or to service providers. There's also an ongoing federal lawsuit over the use of this technology. I'm still looking into it though. edit: looks like the equipment is still under development, experimental, and not commercially available. Phooey. I can't find any development timelines, either, and it looks like the FCC wants to restrict devices using certain frequencies to certain geographic areas, so it could be a while before it becomes possible to roll this out.
    - trotsky 16y ago
      You need an experimental license because none of the (still prototype) devices have received FCC approval yet. In theory, the use of the whitespace frequencies is already open to the public - if you had approved equipment. The rate that the FCC is approving experimental licenses has accelerated a lot in the past 9 months, and new deployments are rolling out much faster as some quasi-commercial gear has started going to trial. While it is difficult to say for sure when the first generally available radios might arrive, I've heard a vendor suggest the second half of 2012. That would be for enterprise gear though, it might take a couple of years before it was hitting consumer price points. The database issues shouldn't be a big deal.
willidiots 16y ago
Whitespace is a mess. Uncertain regulatory restrictions, narrow channels, lack of standardization. That said, it's got a lot of potential for specific uses, as this article shows. The CSMA congestion problem already exists in the WISP world. Ubiquiti built their own in-house TDMA protocol to get around it. I think Alvarion and Canopy have similar, if simpler, timeslotting capabilities. What's interesting is that the 802.11 standard actually includes a Point Coordination Function for this sort of application, but very few (if any) chipsets have ever implemented it.
losethos 16y ago
God says... bought courtly mischief builded commit speeches confine hesitating madness inharmonious detested whithersoever built

43 points by abraham 16y ago

And what does the signal look like if you tune a TV to that channel? Is it possible to fabricate a signal that shows up as a recognizable image?

StavrosK 16y ago
From what I gather, it's already very recognizable as moving specks of salt and pepper.
wiredfool 16y ago
Presumably, it looks like the sky over Chiba city. Maybe with a bit of the sprawl ghosting in it.

derobert 16y ago

"The longest link he could make with existing point-to-point WiFi connections was 400-500 feet; with the new Super WiFi gear in the TV band, he can reach a mile—and it's not a point-to-point signal." Something doesn't make sense here. This is the furthest they could get with point-to-point line of sight links, on towers. When I worked at a WISP, we routinely made longer PtP links—heck, even longer point-to-multipoint links. A mile should be easily doable with a clear line of sight. Using http://www.afar.net/rf-link-budget-calculator/ http://www.afar.net/rf-link-budget-calculator/ shows that a five-mile link should be no problem with 24dBi grid dishes (which are more or less the size of DirecTV, DISH, etc. dishes).

wmf 16y ago
"signals don't easily penetrate the tree branches and leaves which surrounded Aguirre's home" The whole point is non line of sight.
- VladRussian 16y ago
  the phased array WiFi switches have interesting feature - adaptively found best beam direction to your computer may actually be not a straight line from the switch to you, instead it may happen to be a reflected beam.
  - wmf 16y ago
    Other than Vivato (who went broke), is there any shipping outdoor phased-array WISP equipment?

thaumaturgy 16y ago

Huh. Anybody have any idea what the FCC restrictions are on stuff like this?

endian 16y ago
"Guerra is part of a Rice team [...whose...] goal is basic research on providing broadband in the TV "white spaces" — empty channels that the government recently cleared for use by unlicensed Internet providers"
- thaumaturgy 16y ago
  Yeah, I was hoping to find a little bit more in-depth information though. I'm finding reports of a handful of companies (including Google) being selected to be database "administrators" for this, but I can't get clarity on whether such a database is supposed to apply to equipment manufacturers or to service providers. There's also an ongoing federal lawsuit over the use of this technology. I'm still looking into it though. edit: looks like the equipment is still under development, experimental, and not commercially available. Phooey. I can't find any development timelines, either, and it looks like the FCC wants to restrict devices using certain frequencies to certain geographic areas, so it could be a while before it becomes possible to roll this out.
  - trotsky 16y ago
    You need an experimental license because none of the (still prototype) devices have received FCC approval yet. In theory, the use of the whitespace frequencies is already open to the public - if you had approved equipment. The rate that the FCC is approving experimental licenses has accelerated a lot in the past 9 months, and new deployments are rolling out much faster as some quasi-commercial gear has started going to trial. While it is difficult to say for sure when the first generally available radios might arrive, I've heard a vendor suggest the second half of 2012. That would be for enterprise gear though, it might take a couple of years before it was hitting consumer price points. The database issues shouldn't be a big deal.

willidiots 16y ago

Whitespace is a mess. Uncertain regulatory restrictions, narrow channels, lack of standardization. That said, it's got a lot of potential for specific uses, as this article shows. The CSMA congestion problem already exists in the WISP world. Ubiquiti built their own in-house TDMA protocol to get around it. I think Alvarion and Canopy have similar, if simpler, timeslotting capabilities. What's interesting is that the 802.11 standard actually includes a Point Coordination Function for this sort of application, but very few (if any) chipsets have ever implemented it.

losethos 16y ago

God says... bought courtly mischief builded commit speeches confine hesitating madness inharmonious detested whithersoever built