The AREDN Core Team has written a letter to Dick Norton, N6AA, ARRL Southwestern Division Director, who has forwarded it to league headquarter. You can find the letter on our home page news: https://www.arednmesh.org/content/call-action-defense-part-97-allocations
Here is Dick's paraphrased response:
"...I have forwarded your email to the entire Board, which includes the Executive Committee that is responsible for the FCC interface, and the Washington attorney. The board is aligned with your thinking, and is doing whatever it finds possible, but this is a difficult issue..."
We expect the league will offer advice back to us on how we can best lobby the FCC to not proceed with these NPRMs. Stay tuned.
Andre, K6AH
Here is Dick's paraphrased response:
"...I have forwarded your email to the entire Board, which includes the Executive Committee that is responsible for the FCC interface, and the Washington attorney. The board is aligned with your thinking, and is doing whatever it finds possible, but this is a difficult issue..."
We expect the league will offer advice back to us on how we can best lobby the FCC to not proceed with these NPRMs. Stay tuned.
Andre, K6AH
We have 13 sites setup with 9cm links running international 3ghz UBNT gear and Mikrotik routers running IBGP that spans St. Mary, St. Martin, and Assumption parishes in South Louisiana. This project has taken 4 years and nearly 30k to build we have done a lot of begging and deal working to bring this to fruition. This reassignment would wipe us out!! I just hung the last link to close the loop 2 months ago. We don't even have the multiple remote receive nodes built for the repeater yet.
Please add my sites to your list
LEPA No.1 29.691735, -91.192253 to Mr Charlie 29.691129, -91.210020
Mr Charlie 29.691129, -91.210020 to Fairview Hospital 29.686478, -91.261492
Fairview Hospital 29.686478, -91.261492 to Plum St Water Tower 29.696103, -91.308590
Plum St Water Tower 29.696103, -91.308590 to Bayou Sale Water Tower 29.674376, -91.478149
Bayou Sale Water Tower 29.674376, -91.478149 to Cypremort Point VFD 29.723023, -91.856548
Cypremort Point VFD 29.723023, -91.856548 to Jeanerette Water Tower 29.912000, -91.675294
Jeanerette Water Tower 29.912000, -91.675294 to KFRA 29.838173, -91.539546
KFRA 29.838173, -91.539546 to KBZE 29.750937, -91.173722
KBZE 29.750937, -91.173722 to Pierre Part VFD Water Tower 29.959159, -91.212701
Pierre Part VFD Water Tower 29.959159, -91.212701 to Belle Rose Water Tower 30.005462, -91.063497
Belle Rose Water Tower 30.005462, -91.063497 to Napoleonville Water Tower 29.935008, -91.015732
Napoleonville Water Tower 29.935008, -91.015732 to Bayou L'Ourse Water Tower 29.719841, -91.074746
Bayou L'Ourse Water Tower 29.719841, -91.074746 to LEPA No.1 29.691735, -91.192253
Here is a map of our network, The plan was to use 9cm as the network backbone and set up AREDN nodes on 5cm for locals to connect to each pop.
I got AREDN's email about this subject last week and I'm quite concerned. Our group operates 5 - 3 GHz nodes and 4 - 5 GHz nodes, as well as a number of 2.4 GHz nodes in support of several communities and 2 EOC's. The county to our southeast has a county-wide system that relies on even more 3 and 5 GHz links to tie the county together. My concern deepened when I received a forwarded email from our ARRL section manager who relayed that our director had returned from Washington with several other directors and, in his words said "it doesn't look good". So I would suggest it's time to step it up a notch.
I started by creating an info packet with AREDN's "Call to Action" article as the cover, followed by the 2 FCC docket fact sheets, with the last page being my personal response, detailing courses of action we can take. This was distributed at our club's Holiday party recently and will also be handed out at our next monthly meeting. The next step was to get all our nodes on the AREDN map. Done. I'm already an ARRL member but I am encouraging others to join the ranks; there's power in numbers. And give to the grossly underfunded ARRL Spectrum Defense Fund. I then started an email campaign, contacting everyone I could think of in the North Texas area, and even outside this area, who was involved with ham activity on those bands, mesh or otherwise. Ditto for any ham involved with emergency response: eoc's, cert, local fire, the local NWS office, etc. That is still ongoing. Then a fellow ham suggested starting a petition campaign. I have come up with a draft petition that I have attached here. Would AREDN consider hosting a petition on this site that we could send potential signers to, and have AREDN forward the signatures to the ARRL or FCC? Or would it have more clout to independently host a petition drive on a petition web site?
One more course of action is to file a comment to the FCC. They are required to accept comments on pending proposals but I don't know how long that window is open so I wouldn't wait. That web site is https://www.fcc.gov/consumers/guides/how-comment.
I think we need to take this threat seriously and move quickly.
Comment Date: (30 days after Federal Register publication)
So the using calendar days it could be January 16th (earliest). Not sure if it was in the federal register yet.
It is relatively easy to test AREDN devices in this 3GHz. These devices are 5GHz mother boards with a -2GHz converter. Just go into /etc/config/wireless and change the 5Ghz channel to, e.g. 5180 or ch 36 and reboot. Doing a 'save' in basic setup, will restore back to original settings. Setting to ch 36 is 3180 in the proposed relocation space. The command "iw phy phy0 info" will show all the available channels that can be configured, just subtract -2GHz.
I tried this on a 10MHz channel width and a Rocket M3 and a Nanostation M3 connected. HOWEVER, the devices were right on top of each other and there was significant SNR degradation just 4' away. Ouch! It appears these devices have a hardware filter blocking in this space. The manufacturer specs say the freq range is "3370 - 3730 MHz*"
If anyone has a spectrum scope to measure the power levels, we could quantify and consider options, e.g. add an amplifier, modify the filter circuit (assuming there is one found), etc.
If there were to be an easy low cost way to convert these devices, then we'd have a path to convert existing equipment.
Joe AE6XE
I do remember being disappointed when I saw the M3's specified frequency range being so far out of the proposed new allocation and also wondered if that was a hardware limitation. Yes it would be great to see on an analyzer, I'm curious about the linearity that far "out of band", and as you mentioned, the power output. Bear in mind too that 3.1-3.3 GHz is only a suggested recourse per the present proposal. I don't understand why THEY don't just use the 3.1-3-3 and leave the "incumbents" on 3.3=3.55 GHz.
As far as the proposed 5.8-5.9 GHz reallocation, am I reading that we will be forced to share with part 15 users? We know what a disaster that is on 2.4! However, many of the upper U-NII-3 wifi channels seem to be seldom used (yet) and many AREDN links are narrow beam, high gain dishes which might cut through the noise without too much interference or loss in s/n(?) But those super long links, and omni or even sector antennas would probably not work very well on shared spectrum the way they now do on part 97 only spectrum.
If hams can't use their spectrum options, then the only other thing is to use permitted higher power output levels on the common channels with part-97 users (who will have to accept interference per part-97 rules). 50-100 watts output should be enough .... it's not going to be pretty ... or cheap.
I remember in the early days people asking about extending the frequency range and AREDN's answer was that testing showed that output dropped like a rock below 3380. I wonder what sort of filter could fit inside the box and have that kind of sharp cutoff? Maybe it is software?
As far as I know, only AREDN is supporting 3GHz devices in the open source community and this is at the UI layer. There is a check if we are on the Rocket M3 device and then translates to show 3GHz frequencies in the menus, but below that layer, only 2 and 5GHz are known.
However, there is this magical calibration data created unique for each device at manufacture time and stored in flash, called the Atheros Radio Test ART data. This calibration data is loaded from flash and used to download or configure the chip registers. The ath9k driver does do some setup of filters in the chipset radio, but I'm not sure if this is in baseband or in the radio RF blocks. I might give this a try by backing up the ART data, then over-writing from a Rocket M5 to an M3. That would be worth a try. However if it works, the signal quality would not be optimal, using calibration data from a different device.
Joe AE6XE
The PA chip in the one I disassembled is an AWT6283R which has a specified minimum frequency of 3300 MHz
That one is obsolete. The newer ones in the same line have a minimum frequency of 3400 MHz. These amplifiers have matching circuits incorporated in them which simplifies design (50 ohm input SWR < 2) but the matching circuit is relatively narrow band. The input SWR shoots up when you stray outside the specified range of frequencies. Probably something similar happens at the output.
I've seen 5GHz and 3GHz adjacent Rockets show each other in wifi scan numerous times. I have one site, all Rockets, with 3GHz and 5GHz on the outside, with 2GHz in the middle. These are 120 deg sectors as close as you can get them side by side. With RF shielding around the antenna and the Rocket enclosed in a metal box, the bleed through is not detected.
Joe AE6XE