11 Channels

Pretty much by definition, wireless is not a Local Area Network, so comp.dcom.lans.ethernet is not really the appropriate newsgroup for this matter. alt.internet.wireless would be a better newsgroup; I have added that newsgroup into the header.

In any case, the answer is NO, that is not true.

What is true is that each channel transmits on a frequency range centered around a base frequency, and that the frequency drop-off around the defined band is not required to be sharp. There is a power drop off in the channel on either side, more of a drop off for the channels beside those, more yet for the ones beside those.

This leak of signal into the next channel does not make those other channels unusable: it adds noise to those other channels, making them -less- usable than if there were no overlap.

The drop-off is defined such that by the third channel over, the interference is negligable. This situation leads to the following typical allocations, in increasing frequency:

[band 1] - [detectable leak of channel 1] [band 2] - [noticable leak of channel 1] band 3 - [peak of channel 1] [band 4] - [noticable leak of channel 1] [band 5] - [detectable leak of channel 1] [band 6] - [detectable leak of channel 6] [band 7] - [noticable leak of channel 6] band 8 - [peak of channel 6] [band 9] - [noticable leak of channel 6] [band 10] - [detectable leak of channel 6] [band 11] - [detectable leak of channel 11] [band 12] - [noticable leak of channel 11] band 13 - [peak of channel 11] [band 14] - [noticable leak of channel 11] [band 15] - [detectable leak of channel 11]

Someone did a study and found that this allocation was not -necessary- in order to get very good performance. Instead, they found that 4 transmitting channels could be used:

[band 1] - [detectable leak of channel 1] [band 2] - [noticable leak of channel 1] band 3 - [peak of channel 1] [band 4] - [noticable leak of channel 1] + [detectable leak of channel 4] [band 5] - [detectable leak of channel 1] + [noticable leak of channel 4] band 6 - [peak of channel 4] [band 7] - [noticable leak of channel 4] [band 8] - [detectable leak of channel 8] [band 9] - [noticable leak of channel 8] band 10 - [peak of channel 8] [band 11] - [noticable leak of channel 8] + [detectable leak of channel 11] [band 12] - [detectable leak of channel 8] + [noticable leak of channel 11] band 13 - [peak of channel 11] [band 14] - [noticable leak of channel 11] [band 15] - [detectable leak of channel 11]

You could shift the channel 8 activity down to channel 7 and shift the channel 11 activity down to channel 9, and still get very good performance, but there isn't any reason to do this unless you can transmit on channel 12 -- which *is* allowed in some areas of the world. In those areas, you can fit 5 useful channels in 12 allocated channels.

It's more like, "You've asked for your tomato to be put on only

1/11th of the pizza, but we can't slice our tomato that small: we can put the bulk of it on 1/11th of the pizza, but some of it is going to end up on the next 1/11th, and a bit of it will even end up on the part after that. If you insist that none of your toppings overlap, then we can fit in 3 toppings, but if you are comfortable with a tiny bit of topping overlap, hardly even enough to taste, then we can fit in 4 toppings. And if you don't mind topping overlap, we can give you up to 11 different toppings.

I have 13 channels on my linksys WRT54GS (after a firmware update) , does this mean I will get 2 free toppings when I order pizza via domino's website?

Mike

Yeah, but they'd only be from the countries that let you use those frequencies...

snipped-for-privacy@hushmail.com (Walter Roberson) hath wroth:

He left off one important word. There are only 3 _non-overlapping_ channels. These are 1, 6, and 11. Channels in between tend to overlap these channels. DSSS (direct sequence spread spectrum) spectra is about 22Mhz wide. Each FCC channel is 5MHz wide, which means that wi-fi spectra occupies approximately 5 channels.

Intel Wireless Hotspot Deployment Guide ftp://download.intel.com/business/bss/infrastructure/wireless/deployment/hotspot.pdf See Fig 6 and Fig 7 on Page 40.

On Tue, 25 Jul 2006 11:28:17 -0700, Jeff Liebermann wrote in :

That's overly simplistic.

802.11b and 802.11g divide the 2.4 GHz spectrum into 14 overlapping, staggered channels whose center frequencies are 5 megahertz (MHz) apart. It is a common misconception that channels 1, 6 and 11 (and, if available in the regulatory domain, channel 14) do not overlap and those channels (or other sets with similar gaps) can be used so that multiple networks can operate in close proximity without interfering with each other, but this statement is somewhat over-simplified. The 802.11b and 802.11g standards do not specify the width of a channel; rather, they specify the center frequency of the channel and a spectral mask for that channel. The spectral mask for 802.11b requires that the signal be attenuated by at least 30 dB from its peak energy at ±11 MHz from the center frequency, and attenuated by at least 50 dB from its peak energy at ±22 MHz from the center frequency.

Since the spectral mask only defines power output restrictions up to ±22 MHz from the center frequency, it is often assumed that the energy of the channel extends no further than these limits. In reality, if the transmitter is sufficiently powerful, the signal can be quite strong even beyond the ±22 MHz point. Therefore, it is correct to say that channels 1, 6, and 11 overlap. It is more correct to say that, given the separation between channels 1, 6, and 11, the signal on any channel should be sufficiently attenuated to minimally interfere with a transmitter on any other channel. However, this is not universally true; for example, a powerful transmitter on channel 1 can easily overwhelm a weaker transmitter on channel 6. In one lab test, throughput on a file transfer on channel 11 decreased slightly when a similar transfer began on channel 1, indicating that even channels 1 and 11 can interfere with each other to some extent.

That'll come as a great surprise to the great majority of us who use wireless _strictly_ for LANS

However, this is true.

So what did you need alt.internet.wireless to get involved for? That pretty much says it... :-)

"Close enough for this context." (Sorry, I couldn't resist).

(etc...)

This is correct and a serious problem for co-located radios, such as at WISP sites. However, having more than one radio on different channels is a rather unusual situation for the typical home user. Let's play with the numbers and see how it works.

Let's take a typical wireless nightmare situation. Your municipal government has just announced that they will use your tax dollars to provide free wireless to the multitudes by installing an access point on a street light outside your house. You look out the window and there's the antenna on the opposite side of the street, 50ft away. A quick check with Kismet shows that they're on Channel 6, insuring that you're stuck with Ch 1 or Ch 11.

Quiz question: Will the interference generated by the municipal xmitter on Ch 1 or 11 be enough to cause a problem and how badly?

Let's pretend they municipal wireless vendor is clueful and does not install maximum overpower 1 watt radio alligators. Let's say 200mw or

+23dBm. The usual omni antenna has +8dBi gain, but the light pole is well over the head of the nearby home, resulting in perhaps only +4dBi gain at street level. I'll guess about +23dBm + 4dBi = +27dBm EIRP.

At 50ft, the path loss at 2.4GHz is: loss = 36.56 + 20Log(MHz) + 20Log(miles) = 36.56 + 20Log(2400) + 20Log(0.01) = 36.56 + 67.6 - 40 = 64dB

I'll assume a typical access point with a yucky 2dBi rubber ducky antenna. Therefore, the rx signal level on Ch 6 at the receiver input is: +27dBm -64dB +2dBi = -35dBm

The 2nd lobe of the sinx/x spread spectra is guaranteed by an agency of the federal govenment to be -50dB below the center power peak 22MHz on either side, or the middle of ch 1 or ch 11. Therefore, the interference signal at the receiver on Ch 1 or ch 11 will be: -35dBm + -50dB = -85dBm

If the user is trying to use 802.11b at 5.5Mbits/sec the required receiver sensitivity for 10^-5 BER is about -86dBm. The receive level has to be at least 11dB higher (fade margin) at 5.5Mbits/sec or the receiver will be completely unreliable. That means the operating receive signal level will be -75dBm.

-85dBm worth of interference isn't going to do too much to a -74dBm receive level. That's 11 dB of margin. The home owner can safely operate his home network without fear that his tax dollars are being used to construct a municipal jamming system.

Obviously, this is a contrived situation, but I think fairly realistic and common. By moving the interfering transmitter closer to the users receiver, the situation will deteriorate rapidly. Inverse square law says that if I cut the distance in half, the signal level will increase 6dB. Assuming the receive needs a S/N of about 11dB to operate, if the original 50ft distance were reduced to about 14ft, the noise on ch 1 or ch 11 will be sufficient to screw things up rather badly. Same with a substantial increase in xmit power.