Strongest Wireless Channel

one possible explanation is that there is less interference on that channel.

I thought about that but I am new to this and you guys are the the experts here. Thank you for the reply.

Everyones circumstances are different.

Thank you for reporting your findings, it could definitely help someone else.

snipped-for-privacy@yazoo.com hath wroth:

Not really. How much stronger? What hardware? External aftermarket antennas? Home made antennas?

Most very high gain (19-24dBi) antennas just barely have enough bandwidth to cover the 83.5MHz width of the 2.4GHz band. Same with some really tiny ceramic substrate 2.4GHz antennas. Typically, they will have about a 3dB gain variation across the band. However, this is not the case with typical 2dBi vertical coaxial and colinear antennas which have more than adequate bandwidth. There are also ceramic bandpass filters on the antenna connection of most access points and some client adapters to reduce interference from adjacent services. These are not all that flat across the band and do vary with production lots. There is also usually a matching network between the antenna and the PIN diode diversity switch, which can also be frequency sensitive. In other words, there are plenty of components that can cause differences in gain across the band.

For a good clue, search the FCC ID web site for your unspecified model hardware and look at the test report. You'll see that the transmit power varies somewhat across the band. The receiver sensitivity does the same thing, but that's not in the report.

I am using a Senoa PCMCIA card with the 2500 something chipset and a 40 inch omni antenna purchased from Pasadena Wireless. It is mounted on the roof with no structures in the way. The signal I am speaking of is a Linksys router WRT54G and when it's on 6 windows says the signal is low. On channel 11 it goes up to good but mostly very good. I may not make sense but I am telling the truth. Thanks for the input.

Hi, Maybe the antenna is tuned to chnnel 11?

snipped-for-privacy@yazoo.com hath wroth:

A 40 inch long omni antenna is going to have a gain of about 10dBi. That's fairly high gain for such an omni and will probably have a correspondingly narrow bandwidth. I'm too lazy to search for all the possible 40" long antennas carried by WlanParts. Is it one of these?

Well, I got lucky. This one is 39" long:

Unfortunately, the data sheet does not include a gain vs frequency, or VSWR vs frequency graph. Same problem on the Comet Antenna (NCG) web pile.

Ok, so I have to do some guesswork. The VSWR of such an antenna is probably fairly symmetrical across the band in free space. However, when mounted improperly, such as too close to a tower, pipe, wall, etc, the antenna changes resonance, usually going down in resonant frequency. That's backwards for what you're seeing, so that's probably not the problem.

That leaves the various filters, board traces, and matching systems in your Senao wireless device. I can't tell how well those are working without putting the radio on the bench. I have little experience with Senao. However, I can assure you that other vendors products vary substantially across the 83.5Mhz of the 2.4Ghz band. I can possibly dig out some old test data if you're interested.

It also shows up in the test data. See the tx power data for the Senao 2511 at:

in Page 13. Channel Freq dBm Limit Pass/fail 1 2412 15.48 30 PASS 6 2437 15.00 30 PASS 11 2462 14.11 30 PASS Note the 1.3dB difference in tx power output between channels 1 and

1. That's not very much, but that's under ideal conditions, without the resonant effects of the antenna, bandpass filters, matching networks, receiver, etc. This is also probably a hand tweaked unit.

Ok, that eliminates any detuneing of the antenna by nearby metal. Figure on staying about 10 wavelengths (125cm) away from the antenna.

Are you reading the signal strength on the WRT54G status page or on the Senao wireless something or other?

Incidentally, some clients and AP's yield different signal strengths when moving data than when at idle. That's because they like to change data speed with and without data, which causes some differences in indicated signal strength.

Ok, so the mystery Senao device is plugged directly into a Windoze computah. I don't suppose it would be particularly difficult for you to identify what hardware you own?

There are too many variables to determine the culprit. I've seen something similar with home made (or badly made) antennas and equipment. It also happens with very badly built pigtails, connectors, and adapters. Without a bench setup and a pile of test equipment, it's difficult to isolate. The easiest way for you do determine the culprit is to substitute parts and pieces of your equipment and see if it also happens with other access points and clients.

The antenna was advertised as 15db for $66.00. Here is the unit I bought:

It seems to work pretty good compared to rubber duck type I have used. Thanks for all of the input.

BTW, what is the best PCMCIA card? Is a 300mw better than my 200?

Jeff Liebermann wrote:

The specs didn't say that was the case. Here is the link:

snipped-for-privacy@yazoo.com hath wroth:

Hmmm... made by "Pasadena Networks LLC" with no supporting data sheet, patterns or test data. 15dBi in a 40" stick is rather impressive, especially when the 15dBi Comet antenna needs to be 70" long. Most

40" antennas that I've found are only 10dBi gain. Let's just say I'm rather suspicious of this antenna. 300mw is only 1.4dB "better" than 200mw. You probably won't even notice the difference. No comment which card is "best".

Do you think the Comet antenna would be better than what I have?

"Jeff Liebermann" < | | For a good clue, search the FCC ID web site for your unspecified model | hardware and look at the test report. You'll see that the transmit | power varies somewhat across the band. The receiver sensitivity does | the same thing, but that's not in the report.

Is there a link to where this data can be soursed?

Don't forget neighbors. What if he's got a neighbor using channel 11?

snipped-for-privacy@yazoo.com hath wroth:

You might find this comment amusing:

Are you comparing this antenna to others online selling a 15dB omni for considerably less? Note the length of our antenna. It's 70" long. Collinear designed omni directional antennas accumulate gain by stacking 13cm segments. There's no way an antenna 30" less than ours can provide 15dB of gain. So buyer beware, not all 15dB antennas sold online are equal, some are actually only 11dB of gain. I rest my case. You really have a 10dBi antenna with little clue as to what it really does in the way of gain, vertical radiation angle, and apparently bandwidth.

Better for what purpose? I'm not a big fan of very high gain omni antennas. The problem is the vertical radiation angle. The Comet

12dBi omni has a vertical radiation angle of 5 degrees. See data sheet at: 5 degress is very narrow. The 15dBi antenna will probably have a 3 degree vertial beamwidth. If you're lucky, these antennas will also have a 3-5 degree downtilt, so that you don't end up talking to the sky (above the horizon). However, that adds length or reduces gain depending on method used to obtain downtilt.

So, your rooftop 15dBi antenna will talk to anything that's within + or - 1.5degrees of horizontal. To put that in perspective, if you mount the antenna on a 20ft roof top, the closest you'll be able to talk reliably is:

0.36 miles or 1900 ft away of no downtilt. For 3 degress beamwidth and 3 degrees downtilt, it's 0.12 miles. However, your maximum range is only 0.36 miles which really sucks. Change the downtilt to 2 degress and you maximum range now goes to a not so fabulous 1 mile. Yech. That's nice for the neighbors, but doesn't do much good for your own use. It's incredibly sensitive to radiation angle and tilt. Just draw a 3 degree angle on a piece of paper to see how small it is. In addition, unless you have very flat terrain, it's going to be difficult to find a location where you're inside the 3 degree beamwidth. That's why I think very high gain omni's are a bad idea.

"NotMe" hath wroth:

Well, he didn't bother answering my questions as to what specific equipment he was using so I'm not 100% sure I have the correct hardware. All I know is that it's a WRT54G and a Senao PCMCIA card. I posted a link to the Senao tx power variation elsewhere in this thread. Quoting myself:

It also shows up in the test data. See the tx power data for the Senao 2511 at:

in Page 13. Channel Freq dBm Limit Pass/fail 1 2412 15.48 30 PASS 6 2437 15.00 30 PASS 11 2462 14.11 30 PASS Note the 1.3dB difference in tx power output between channels 1 and

1. That's not very much, but that's under ideal conditions, without the resonant effects of the antenna, bandpass filters, matching networks, receiver, etc. This is also probably a hand tweaked unit.

I'm too lazy to lookup the WRT54G, but I suspect the variations in tx power will be about the same 1-2 dB. That's not much, but remember that these are hand tweaked units. Production numbers are considerably worse.

However, in the OP's case, I don't think the problem is with the radios (tx or rx). My guess(tm) is that he as a rather narrow band antenna, which apparently is also lacking in any published specifications. High gain antennas can just barely squeeze by while covering the entire 83.5MHz of the 2.4GHz band. Most of the really high gain units have a marginal 2:1 VSWR at the band edges. Any variations in tuning caused by nearby metal or production variations, will cause drastic VSWR changes. The mismatch will also mis-terminate the ceramic filters inside the unit, which will add their own variations in VSWR and loss.

Tim Smith hath wroth:

Sure. Someone else mentioned interference. Interference will affect the noise level, but not the signal strength. However, the OP ignored my request for numbers and instead supplied the "number of bars" which could mean anything including signal to noise ratio and bit error rate, which will certainly be affected by interference.

My experience with interference is that it doesn't stick around. If the OP had mentioned that it *SOMETIMES* is "stronger" on ch 6 than ch

11, I would certainly suggest interference. He didn't, and my crystal ball suggests that he's also done some tinkering, which implies that it's probably a permanent condition. Interference is also position sensitive. He's using a laptop which in the presence of interference, will have variations in interference levels. Unless the interference source is in the same room, methinks that's not the cause.

I just thought of something. What fequency are Uniden 2.4 gig phones on? I want to stay away from that channel, correct? TIA

snipped-for-privacy@yazoo.com hath wroth:

Uniden uses 4 different chipsets in their various 2.4GHz phones. Each one works slightly differently. Some select a specific 1Mhz channel from among 80 available and use ordinary FM modulation. Some frequency hop over the entire 2.4GHz band. One uses direct sequence spread spectrum, but not exactly in the same way as 802.11. The latest are OFDM. All of them will interfere with your 802.11b/g wireless because they use exactly the same frequency band going from

2400 to 2483.5MHz. There are also some that use 2.4GHz in one direction, and 5.8GHz in the other. I think there's also a 2.4GHz/900MHz version, but I haven't seen those lately.

If your Uniden cordless phone uses direct sequence modulation, it's smart enough to sniff the band for traffic and look for an empty area. Usually, the cordless phone starts at the bottom of the band. Therefore, if you put you 802.11b/g wireless on the highest channel

11, you have a fair chance at avoiding the interference. Otherwise, you lose.

I presume you have a Uniden phone. As with your other wireless equipment, you seem to have an inability to disclose the exact model numbers. Probably caused by too much RF exposure. If you want help with this, you'll need to supply the Uniden model number. If you want to make it easy for me to do the necessary chipset investigation, also supply the FCC ID from the serial number tag.

The reason I didn't mention the model is because I am not sure because the sticker is old/faded but it looks like it could be exi3246 and the phones are marked 2.4ghz. Thank you.