Author Topic: The sound card side of PDW must get better  (Read 15625 times)

genobpgh

  • Member
  • *
  • Posts: 6
    • View Profile
The sound card side of PDW must get better
« on: 27 Jul 2014, 15:01:37 »
I'm really happy to finally get to see the PDW source.  It answered a lot of questions to why it was hit or miss when you use a sound card as the input vs using a hardware slicer.

Since most of the paging in my area is FLEX, most of my work/comments will focus on that protocol.

Considering the bit-rates of FLEX, there is NO REASON you should not be able to use the sound card in your computer.  If you have a decent radio (most radios capable of 900+mhz are good) you should be able to get a signal through audio without using an IF tap (unless the audio out is being filtered < 12khz).

Now I am generalizing a bit.  But - I've been working in signal processing since the 90's (especially with sound cards) and I can tell you that the A/D's have made leaps and bounds in signal quality.

Observations over the years:

1) PDW can only resolve a percentage of messages that come through.  Why?
2) The code samples at 8bit - all of the math is in 8bit.  Time to upgrade to 16 or higher if you have a 24bit sound card.
3) The data slicers work much better than the sound card.  There is no reason the sound card should be inferior in the year 2014 :)
4) The FLEX decoding seems to be missing a few SYNC frame types.  We should resolve that.
5) After how many years of FLEX pagaing, should a spec have leaked out that can really explain the protocol.  OR are we still reverse engineering signals?
6) Currently most of the flex signals I'm seeing these days are not being decoded.
7) The current decoding code looks like it was written in the 90's and ported up to windows at some point (is that true?).

What I plan on doing:

If someone can come up with a legitimate FLEX decoder datasheet - or something that explains all of the FLEX/REFLEX sync headers that would be GREAT!

I am currently working on a 16bit module that will detect/decode FLEX.  I'm basing it on a bunch of information I found on the internet and the PDW code source.  If I can get it working and decoding I will let this development forum know and hopefully we can integrate a more complete and reliable 16bit FLEX decoding engine into PDW.  I plan on open sourcing everything.  I just want the sound card side of PDW to work much better than it currently does.

Thanks for listening!

genobpgh






Discriminator

  • Administrator
  • Top Member
  • ***
  • Posts: 185
    • View Profile
Re: The sound card side of PDW must get better
« Reply #1 on: 02 Aug 2014, 08:41:27 »
The data slicers work much better than the sound card.  There is no reason the sound card should be inferior in the year 2014 :)
I fully agree. It should also be possible to decode 4-level modes with the soundcard. The PDW community will be happy to see improved soundcard performance!  :)

RFBUG

  • Member
  • *
  • Posts: 2
    • View Profile
Re: The sound card side of PDW must get better
« Reply #2 on: 22 Sep 2015, 10:56:06 »
I know this is an old post.
 I'm assuming you are talking about using a speaker / ext speaker output on a radio receiver, while what you say has merit in making it easier to set up to monitor paging and other digital signals transmitted over radio waves, this type of audio will never yield the best results as this audio is for the human ear and is filtered and shaped to make listening not so harsh on us humans, there are very few radio receivers on the market that have a raw audio output for this purpose, and the same goes for transceivers as well.
To get the best results with a radio receiver that doesn't have a raw audio output is to fit a IF/DISC tap.

Another common mistake by some that use speaker audio is make sure the squelch is off as this will cut the beginning of data if it is adjusted to cut static when no transmission is present.

When I first started using PDW I to first used speaker audio and did not have the best results, since I fitted a DISC tap and I do use my sound card I have had very good reliable decodes with the very very odd corrupted message, but that happens.

Discriminator

  • Administrator
  • Top Member
  • ***
  • Posts: 185
    • View Profile
Re: The sound card side of PDW must get better
« Reply #3 on: 22 Sep 2015, 18:25:02 »
To get the best results with a radio receiver that doesn't have a raw audio output is to fit a IF/DISC tap.

Yes. Did you check the domain name of this forum? discriminator.nl -

'If you want to monitor digital modes that use FSK or PSK (e.g. POCSAG, FLEX, ERMES and AIS), you get the best results if you use the unfiltered audio from your receiver. You can find that signal at the discriminator of your receiver. A discriminator is the 'heart' of an FM receiver. After the discriminator, audio filters and amplifiers follow that may heavily distort a digital signal. If you want to decode and monitor digital (FSK and PSK) systems seriously, a discriminator tap is an absolute prerequisite. '