Whitespace wifi

[Modelworks]

The FCC has selected Wilmington, NC as the test city for whitespace wifi. I live about 50 miles from there so my interest peaked. The idea is to use frequencies that are not in use but already assigned in the overall allocations. Example would be if the frequency for channel 22 on your tv is not used locally that it could be used for other purposes. Speeds are lower than traditional wifi right now but it shows potential. The lower frequencies mean that it can penetrate walls, trees, buildings easily compared to 2.4ghz+ .

I see this as a plus and minus. It is good for people wanting to make a connection but bad in that your signal is now traveling much much further. Interested in what others think ?

I thought I should clarify what 'much' means. A router using a radio now might keep the signal within a home, or 100ft or so indoors. Whitespace would change that so the same broadcast power would now reach 2-3 miles.

 

[drebo]

This is absolutely amazing. This is what WISPs need to become more competitive with wireline providers.

Right now, we're using 3.65ghz and 5.8ghz and we have customers as far as 10 miles out. However, they require line of sight. Trees, buildings, even birds can sometimes cause issues.

Down in the 700mhz range, the effects of obstructions are much, much lower.

However, it will really depend on how the FCC handles this. THe way they've handled 3.65ghz isn't really all that great. It could be much better. If they do it in a similar way (pseudo-registered) but do not require registration of client radios, and open more than 50mhz of bandwidth, it could make for some very interesting times. Transmit power will be interesting, but we'll have to see.

I've been wanting use of the old TV whitespace for a long time. It's great to see that we might actually get it. 5.8ghz and 3.65ghz are OK...I get 30mbps at 3 miles out from relatively weak PtMP connections...but I would prefer further coverage. Most of my customers are <5mbps so 10mhz channel widths wouldn't bother me too much for frequency coordination purposes. I just need dependable, low-latency, reasonably priced, and regulatory-easy devices.

 

[Modelworks]

However, it will really depend on how the FCC handles this.

They have a couple different options for use of the frequencies. There is fixed point, like a wisp would use and there is also home use. The main difference between the two is home users are not allowed to use antennas to broadcast in an omni-directional way for their home, fixed point users can broadcast with an antenna to other locations for point to point, great for wisp use. Users are limited to 100mw max That gets you a lot of distance. 100mw means at 530Mhz I can cover 2.3 miles where I live with the standard soho router antenna setup. My main concern with the unlicensed part is that home users are going to be much more likely to interfere with each other.

If you want to know what you can use in your area for frequencies the company spectrum bridge maintains a list, just put in your zip code and it will tell you what the fcc has allowed for your area.
http://whitespaces.spectrumbridge.com/whitespaces/home.aspx

For me it shows this for home use:
channel frequency power in mw
21 512-518 100
22 518-524 40
24 530-536 40
25 536-542 100
26 542-548 40
30 566-572 40
33 584-590 40
40 626-632 100
41 632-638 40
45 656-662 40
50 686-692 40
51 692-698 100


If I switch it to fixed point:
channel frequency antenna height in meters allowed.
2 54-60 30
5 76-82 30
6 82-88 30
7 174-180 30
8 180-186 30
14 470-476 30
15 476-482 30
21 512-518 30
25 536-542 3
40 626-632 30
51 692-698 10


Something to understand about whitespace is height of the transmitter factors into it too because the signal from the tv transmitters gets lower to the ground as it gets farther from the transmitter. You can still use a frequency that a distant tv station is using but your antenna has to be low enough that the signal of your transmission falls off before it enters too far into the tv stations broadcast area.

Above it shows 25 as allowing only a 3 meter height for antenna because there is a station about 45 miles away that also uses that frequency.

The main obstacle right now is the cost of the radios, some want 20K for a set which is outrageous considering there is nothing costly about them over traditional wifi type radios. It doesn't require any special technology to implement that hasn't been around for 30 years. The FCC even dropped the requirement for detection of stations in use for the area which was really the main cost of developing the system .

Myself I plan on building my own set just to test it out, speeds will be low since I plan to just stick to a single frequency but it could scale up to multiple bands for more throughput.