[HamWAN PSDR] 902 MHz Band Plan? Hardware?

Ryan Elliott Turner ryan.e.t at gmail.com
Fri Nov 21 16:46:11 PST 2014

"Nv2 is supported for Atheros 802.11n chips and legacy 802.11a/b/g chips
starting from AR5212, but not supported on older AR5211 and AR5210 chips.
This means that both - 11n and legacy devices can participate in the same
network and it is not required to upgrade hardware to implement Nv2 in

Based off of that statement, the XR9's AR5414 chipset should be compatible
with NV2. That'd certainly be interesting :)

As for centers/sizes that are abnormal, the mikrotik feature supporting
that is "Advanced Channels"; here's a snippet on that:
"Currently only Atheros AR92xx based chips support non-standard center
frequencies and widths with the following ranges:
center frequency range: 2200MHz-2500MHz with step 0.5MHz (500KHz), width
range: 2.5MHz-30MHz width step 0.5MHz (500KHz);
center frequency range: 4800MHz-6100MHz with step 0.5MHz (500KHz), width
range: 2.5MHz-30MHz width step 0.5MHz (500KHz);"

Corresponding with that, the 9HPn's radio is AR9283-AL1A.

So, based off of that info, I believe yes, the XR9 will run NV2 on a
mikrotik RB, and no, I don't believe it will support advanced channels' fun
step/width capabilities. The advertised widths for the XR9 are 5, 10, 20,
and 40.

I think that the Mikrotik XR9 in a MikroTik RouterBOARD RB/411GL would
probably be a good solution.

Regarding sectorization...
I believe that we should be OK to use at least those two channels. The
third I'd believe to be safest would be 903-908, fully avoiding the newer
25 MHz spacing FM repeater pairs and FM call freqs. Basically, then we'd be
overlapping CW, beacons, and old channels. Still suboptimal, still likely
to encounter interference.

Let's remember though that the added complexity and cost of the XR9 + RB411
only gains dual pol; if we anticipate users are going to be single pol on
whip antennas, what's the point? I see this not only as a mobile solution,
but also as a NLOS solution for those "last hope" situations.

On Fri, Nov 21, 2014 at 5:59 PM, Bart Kus <me at bartk.us> wrote:

>  Are you sure a Mikrotik host board won't just run a 902MHz card like the
> XR9?  If it's possible, that'd allow us to keep the same software stack we
> have right now, keep TDMA, and provide the MIMO we're seeking.  The other
> thing I'm curious about is the frequency agility of the XR9.  Does it
> support 2.5MHz/5MHz channels?  What are the quantization steps on its
> tuning ability?  What is its tuning range?  Are there other 902MHz
> miniPCI(-e) cards on the market?
> I'd still like to see sectorization since it's hard to scale otherwise.
> The band is 26MHz wide, which is enough for 5MHz sectorization.  Perhaps
> the sectorized spectrum can be arranged to avoid the most likely/frequent
> users of this band, but I don't think you're gonna have any luck by trying
> to comply 100%.
> If someone has any suggestions about other technologies, such as CDMA
> spectrum sharing, I'd like to hear them.  One of the things to keep in mind
> about 902MHz HamWAN service is that it's aimed squarely at mobile users.
> Can't assume high gain directional antennas at the user terminal.  I'd like
> to support things like whips on car rooftops with near-continuous coverage
> as the car drives around.
> As some of you are aware, I'm not satisfied with the available power
> levels for any of this equipment.  Mobile stations will be behind buildings
> and trees, at close to ground level.  I'd like to bring things up +20dB or
> so over stock (ie: 50W range).  If you know of any affordable 902MHz
> amplifiers that can be used with these radios, let me know.  The
> requirements are RF-sensed TX triggering from signals as low as 100mW or
> so, bidirectional signal handling, about a 1us TX/RX delay time, and linear
> amplification since these modems use amplitude components to carry data.
> Running off a standard 12V car power system would be a bonus.  The
> combination of reduced frequency and increased power ought to drive a far
> more available service than we have with the 6GHz offering right now.
> However, experiments can begin right now with just stock power levels to
> get a better feel for propagation properties and noise levels of this band.
> --Bart
> On 11/21/2014 10:34 AM, Ryan Elliott Turner wrote:
> Hey folks,
>  I'd love to get rolling on testing some 900 MHz gear here in Memphis.
> I've got a few high-level questions about it, though, and I'd love to find
> out what others have already settled.
>  Do we want to do sectors and multiple channels? UBNT sells a AM-9M13
> that looks pretty slick; 120 degrees, 2x2 mimo; spec sheet here:
> http://www.streakwave.com/mmSWAVE1/Video/AirMax900Sector_DS.pdf
>  Mikrotik sells a metal 9hpn that's 27 dBm @ 11Mbps to 24dBm @ 54 Mbps
> (500 mW transmitter), but that's single chain. From what I read, dual chain
> MIMO mitigates about 3dBm of noise, which may be worthwhile. These devices
> don't play nice between vendors, though, and I think the preferred dual
> chain 900 mhz device is the UBNT XR9. We'd lose TDMA by going with this but
> gain another chain. Thoughts?
>  Apparently the ARRL band plan is pretty outdated and unrecognized (
> http://www.aracc.org/33cm900mhz.pdf ). Around here, we have an
> organization named SERA that has specified a 900 MHz band plan; it is
> available at the following URL:
> http://www.sera.org/pdf-files/SERA%20902%20MHz%20FUP.pdf
> So, here we've got
>    - 922<=f<928
>    - 910<=f<916 (it says ATV...)
> And that's it. So, if we were to sectorize, we don't even have 3 channels
> to use here.
>  Thoughts? Looks to me like omni, TDMA, and single 5 MHz channel
> (911-916?) is the best way to go for this.
> --
> Ryan Turner
> _______________________________________________
> PSDR mailing listPSDR at hamwan.orghttp://mail.hamwan.org/mailman/listinfo/psdr_hamwan.org
> _______________________________________________
> PSDR mailing list
> PSDR at hamwan.org
> http://mail.hamwan.org/mailman/listinfo/psdr_hamwan.org


Ryan Turner
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.hamwan.net/pipermail/psdr/attachments/20141121/6b58c8e7/attachment.html>

More information about the PSDR mailing list