[chbot] Planning for the next few weeks - SDR Experts?

[Mark Atherton]

I would start off with a PC and a dongle capable of a waterfall display 
centred on the nominal carrier, with scan width set to maybe 3kHz, then 
you may be able to find the carrier.

Then you can improve centre frequency so you can drop scan width, and IF 
bandwidth.

Repeat until carrier is within a couple of Hz, and IF is maybe 20Hz 
wide. You now have the possibility of a very sensitive receiver.

If you can get an SDR handheld with an IF bandwidth of 20Hz, all you 
have to do is transfer the exact frequency of the carrier, and you have 
a very sensitive portable device.

In terms of automation, initially don't. Get the hang of a process that 
works, and equipment that will do the job. Once you have ironed out the 
fiddling to make the system work, then you can transfer your knowledge 
into an algorithm to automate the process.

Required laptop and SDR sounds easy to set up. Handheld SDR sounds 
expensive / difficult.

Having said that, you could set up a network of laptops + SDR dongles at 
fixed locations and manage some kind of tri/quad/hex-angulation. Each 
dongle must have an accurate and stable timebase. Laptops must be 
connected to the same network.

Even a 50Hz wide IF for a CW receiver might do the job, but you will 
need around 10Hz accuracy at 300Mhz which 33 x 10-9 which sounds like a 
Rubidium reference standard.

The other option is to load a calibration number into the handheld each 
day to adjust it's reference which will hopefully stay put for 24 hours.

Just wondering about a homebrew Zero-IF RX, you can easily crank down IF 
bandwidth on such a device...

Hopefully some of that makes sense.

-mark



On 22/03/2020 1:22 PM, Trevor Wignall wrote:
> The frequency is around 300MHz and is known to within a couple of kHz 
> - later models are a bit more stable, so perhaps within +/1 1kHz. 
> Repetition is 1-2 seconds. So essentially a CW (morse) transmitter 
> sending the word "e". Transmit power and aerial are not relevant to 
> trying to improve the receiver - I'll take any improvement I can get! 
> There is a test transmitter operating with a coverage of most of 
> Christchurch - clearly heard at the Idris Rd clubrooms.
> I don't want to put too many specifics in an open thread - anyone 
> seriously interested can contact me and I will give them more 
> specifics (after I have cleared it with the Wandersearch Trustees).
> Trevor
>
>
> On Sun, Mar 22, 2020 at 12:59 PM Mark Atherton <markaren1 at xtra.co.nz 
> <mailto:markaren1 at xtra.co.nz>> wrote:
>
>     Some parameters please
>     - max and min limit of carrier frequency, or Fnom ± Ftolerance
>     - repetition rate of the 100ms bursts ?
>     - transmit power
>     - transmit antenna
>
>     Do you have a pendant in hand to take apart ?
>
>     So unmodulated carrier, and no ident modulation ?
>
>     - Mark
>
>
>
>     On 22/03/2020 12:44 PM, Trevor Wignall wrote:
>>     Hi all
>>     As a Wandersearch responder (i.e. someone who goes out with a
>>     directional receiver and tries to locate the small transmitter
>>     pendants worn by people who tend to go wandering e.g. dementia
>>     patients), I am looking at using a Software Defined Radio (SDR)
>>     to receive these signals. However as the signals are weak, I am
>>     looking for ways of extracting the short transmitted pulses
>>     (around 100-200 milliseconds of unmodulated carrier) from the
>>     background noise using narrow band filtering. The problem is that
>>     the exact transmitter frequency is not known due to tolerances,
>>     temperature changes etc.
>>     If anyone familiar with SDRs would like a challenge to combat
>>     boredom, any help would be appreciated.
>>     Trevor
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ourshack.com/pipermail/chchrobotics/attachments/20200322/d2358deb/attachment-0001.html>