Association list signal/noise readings

[bobbyc]

Hi, Is the signal and noise readings reported from the customer (what the customer is hearing from the hilltop), or the hilltops readings (what the hilltop is hearing) I'm assuming it's the customers, because if it was the hilltops, the noise floor would be the same per customer?
Thanks, Bob C

[tony]

Hi, Is the signal and noise readings reported from the customer (what the customer is hearing from the hilltop), or the hilltops readings (what the hilltop is hearing) I'm assuming it's the customers, because if it was the hilltops, the noise floor would be the same per customer?
Thanks, Bob C

Yes, it is what the customer is hearing from the hilltop.

Thanks!

[bairdc]

I'm assuming it's the customers, because if it was the hilltops, the noise floor would be the same per customer?


Don't be so sure about the noise floor being the same for everyone at the AP. I would have expected that to be the case as well, but I've got an interesting situation on one of my APs. Somebody recently "turned something on" near this AP that has now resulted in a noise floor of about -75 (Ugh!) around the clock. However, I've got a couple of customers that are quite close to that AP, and are running fairly high gain dishes (21dB). StarOS (the AP) quite often shows a noise level of -96 for both of these customers. Both of them do fluctuate between about -75 and -98, but all of my other customers who are not so close seem to hang right around -75, and never get even close to the -90 range. I don't have any explaination for this. I would expect the noise floor to pretty much be the same for all customers at the AP. But apparently that's not always so. Anybody have an idea why this is? I'm curious...

Craig

[bobbyc]

That doesn't make sense to me, the noise floor of the tower would be consistent... it's the noise floor for that specific card...
Tony, is the reason the noise floors are different on the association list because each customer doesn't update at the same time, and my noise floor is flucuating?
Thanks, Bob C

[georgew]

There are wireless service providers selling cable-TV like services that operate above and below our 2.4ghz band. These companies are going out of business at the moment because satellite and wired-cable companies are offering more channels.

These wireless cable operastors are starting to sell broadband access as a replacement for the lost TV revenue. These companies have huge investments in their radio licenses, and they see us as maggots eating into the value of their licenses... much like the phone company feels about wired ISP's...

The wireless cable operators are very much aware of our networks, and all they have to do to shut us down is turn some knobs on their transmitters, causing them to totally blank out the 2.4ghz band with noise.

What is worse, they are doing it on purpose!! I repeat, these guys are purposely stuffing our 2.4ghz band with noise to shut us down, and to allow themselves to attack our customerbase.

I know this because one of the operators of one of these companies told me about the practice. He says that on average it takes the FCC 3 to 6 months to tell them to stop, and in some markets the complaint never comes. If someone has suddenly "turned on" a noise source, it would be in your interest to find out exactly who and what it is, so you can quickly register your complaint with the FCC. It may be your local wireless cable operator trying to steal your customers by ruining your service.

These guys need to go out of business, and the FCC needs to reallocate their bandwidth to us. The way they work, they have countless technical problems that would be solved by using our methods, but it isn't going to happen untill the last of them are gone.

In any event, it would be a good thing if we developed a team approach to this problem, because I know that as individuals few of us can afford the equiopment or knowledge to combat these companies. If we had an association of wireless operators teamed together, we might be able as a group to buy the kind of gear we need to track down and stop these attacks. We would have to figure out how to share the equipment so we all get a chance to use it when we need it, but this would solve the problem of all of us having to buy expensive spectrum analyzers to solve noise problems and noise attacks.

Essentially this is a kind of denial of service, and if we started proving the abuse the feds would probably take care of the abusers for us.

A small group of us could buy a spectrum analyzer, and share it's use by mailing it from one person to the next on some sort of fair schedule.

I know many radio drivers have spectrum analyzer-like features, but so far I am unimpressed with them, and they do nothing to solve problems like this...




George

[bairdc]

I know it's not a wireless cable operator doing this. We don't have any in our town. However, there are a couple of nearby cell towers. I'm suspecting one of them has slapped in a WMUX lynx radio or something for backhaul. I'll be heading up there soon to see if I can locate the source of the interferrence.

[bobbyc]

Our electrical company has a OLDER WMUX 2.4 Circularly polarized link on the same hilltop as us... so we get local interference plus their uplink antenna is about 3 miles away blasting right at our sector!
On the tower, to minimize local interference, we have OEM agere w/out internal antennas, and our radiowaves horizontal sectors have metal side-shields folded in to protect the sidelobes of the sector.
Unfortunately, that sector is amped, since our customer base 8 miles away. We are amplifing a lot of noise. The noise level is typically -84.
What are others seeing their noise level at on amped towers?
Bob C

[bairdc]

That doesn't make sense to me, the noise floor of the tower would be consistent... it's the noise floor for that specific card...
Tony, is the reason the noise floors are different on the association list because each customer doesn't update at the same time, and my noise floor is flucuating?
Thanks, Bob C

And here's something that confuses me even more: This interferrence was on channel 11. When it popped up, most of my customers' associations were showing up in StarOS with a quality of 5 or below. In fact one or two were at 1 on quality. However, for the most part, they were all still running at 5.5 or 11 megs, and seemed to have solid links. There were no customer complaints either. But in an attempt to try to solve the problem, I switched to channel 8. I happen to have another antenna on this same tower using channel 6, but I figured I'd give it a try anyway. In spite of the fact that I know my other AP was interferring with this one, my noise floor went up to around -84. It did fluctuate a lot, but it was significantly better overall, and most had quality of 10 or better after the switch. However, believe it or not, after doing this most of my users dropped to 1 meg, and I got a couple of complaints from gamers about packet loss. So how is it that I can improve my noise floor, thereby improving my quality, yet have poorer performance? Weird! :?

Craig

[bobbyc]

I must ask again; what are you fellas seeing for a noise floor on your amped sectors?
Thanks, Bob C

[dkii]

I don't think this would be too widespread of a problem, since I don't think there are many of them out, but I was talking with tech support at breezecom trying to resolve some performance issues, and he asked me if there were any XM terrestial repeaters around. Turns out these guys are at(I believe) 2.35ghz, and are allowed by the FCC to xmit up to 40kw. He said most of them are around 2kw, and they have had problems with their wireless equipment in those areas. He advised that I purchase a bandpass filter and place it in line with my antenna cable and this would eliminate any possible out-of-band interference like the XM repeaters. I don't have an XM repeater in my area that I know of, so that wasn't my problem, but it may be a problem for some of you, and either way I think the bandpass filter is a good idea. Just ftr, I was playing with my test soekris unit and noticed on chan. 11 a 75db noise floor, I don't recall if it stuck around, but this had no external antenna connected. It was on one of the lucent cards with the integrated antenna. I'll check on it later and see if its still there.