Atheros DFS radar detection

[Adrian Chadd]

Hi,

Just as a reference, I've committed the strong signal diversity
twiddle to -HEAD as well as AR5413 radar parameter support (in case
you come across one of those.)

That completes my (current) round of help with this, at least until I
get back from my trip.

Of course, it'd be really great if someone signed up for working on a
BSD licenced radar pattern detector.. :)



Adrian

On 26 August 2012 22:14, Wright, Brett
<Brett.Wright at cooperindustries.com> wrote:
> Hi Adrian,
>
>> -----Original Message-----
>> From: adrian.chadd at gmail.com [mailto:adrian.chadd at gmail.com] On Behalf
>> Of Adrian Chadd
>> Sent: Saturday, 25 August 2012 3:52 AM
>>
>> I haven't yet sat down and done actual radar pulse train detection
>> with the AR5212 and I know it's just going to plain suck at doing
>> chirps in a busy environment (as they just look like long pulses;
>> there's no FFT data to assist the software in decoding what's going
>> on) but it should be enough to get you started.
>>
> That might be half my problem, the radar patterns I have to setup to get
> compliant detection increase the chance of false radar train detection
> (in combination with the false pulse detection). Reducing this false
> pulse detection has been the biggest help. If I tinker too much with the
> radar patterns I can't get compliant detection.
>
>> Have you tried these NICs on different hardware? I wonder if you're
>> seeing some kind of spur or board noise that's contributing to both
>> the false detections and the high noise floor calibration value. I
>> don't have an SR-5 here to test against I'm afraid..
>>
>
> I haven't tried other hardware because I'm working on an existing
> hardware platform where I have to use SR5 (mainly for the high Tx power
> etc). But makes sense that the SR5 is making this whole process more
> difficult...
>
> Thanks for all the help.
> Brett