[pskmail] Re: Various subjects

  • From: John Douyere <vk2eta@xxxxxxxxx>
  • To: pskmail@xxxxxxxxxxxxx
  • Date: Fri, 8 Jan 2010 07:00:05 +1100

Hi Per,

Here is the information on the interleavers:

To maintain constant time delay between bits, we double the number of
concatenated square interleavers for each doubling of speed:
2x2x40=160 for PSK125R, 2x2x80=320 for PSK250R, etc..

On the RSID side I haven't looked as the one you mentioned yet. but I
have been thinking about a more reliable way to id the modes.

In fact I have been wondering if there was not a way of doing direct
id using the preamble pattern for a limited number of modes. The
signature of the preamble is quite distinctive between the bpsk, pskr
and thor modes for example. I will discuss this with Dave to see how
practical it is.

The alternative is a more robust RSID based on the most robust mode we
plan to use, like THOR8 for example. We could then pass on more
information than just the mode. And the advantage of a "reference
mode" is that the receiving Fldigi knows which mode to go to by
default. Trimming preamble and postamble could produce an acceptable
Tx time.

Or modify the current rsid (rsid2) to make it more robust and maybe
include extra information.

A few options. But I agree we have to get the rsid decoding more reliable.

I will make this my task when I come back mid next week.



On Thu, Jan 7, 2010 at 8:20 PM, Pär Crusefalk <per@xxxxxxxxxxxx> wrote:
> Hi John,
> For the user guide/manual I would like to add a simple table showing all
> the modes with their respective characteristics and a few hints about
> how to use them best. I know we discussed the inclusion of an
> interleaver and fec for the robust modes but I'm not sure where you
> landed with regards to the size etc. Would you mind filling in the
> blanks if I send you a table showing the modes available?
> While on the subject of new modes I wonder if you have looked any at
> MIL-STD-188-110 and the autobaud feature? In short its a better RSID as
> its in PSK and is sent at the start of each frame. It details speed and
> interleaver size etc. In general I think the idea of a mode id is the
> right one but perhaps it is already done (and better). RS ID is nice and
> all but its not always received and understood.
> Making it possible to adapt the interleaver size could help avoid fade
> trouble (short interleaver is then better as not all frames get
> scrambled).
> 73 de Per, sm0rwo

Other related posts: