RTTY
[Top] [All Lists]

Re: [RTTY] bartg sprint 75 sv1dpi

To: Peter Laws <plaws@plaws.net>, RTTY <rtty@contesting.com>
Subject: Re: [RTTY] bartg sprint 75 sv1dpi
From: "Joe Subich, W4TV" <lists@subich.com>
Date: Mon, 19 Sep 2011 18:04:12 -0400
List-post: <rtty@contesting.com">mailto:rtty@contesting.com>
 > I'm guessing that what I really need is a 75-baud "profile" in MMTTY
 > that optimizes things for that speed (along the lines of 23-Hz shift
 > profile that's distributed with the software).

The 45.45 baud ("Standard RTTY") profile appears to work fairly well
except for the Pre-Filter bandwidth.  I have changed the 45.45 baud
Pre-Filter settings to Taps=192 and FW=100 for a flatter passband and
sharper skirts.  For 75 baud, I have been experimenting with Taps=256
and FW=150 to avoid the "too narrow" issues with 75 baud.

The "stock" bandwidth values for the IIR and FIR resonators in the
discriminator appear to be reasonable for either speed but I have
not attempted to generate data signals with measured S/N ratios and
process them with with alternative resonator parameters.

The bigger issue is the use of overly narrow transceiver filtering -
either crystal or DSP filtering - particularly when combined with
"Dual Peak" filters optimized for 45 baud operation.  The formulas
in NTIA Annex J give 250 Hz as the minimum bandwidth for 45.45 baud
FSK (1.2 x shift + baud rate) which makes the popular cascaded 250
Hz filters very tight for even 45.45 baud (250 Hz is between -6dB
points).  For 75 baud the minimum bandwidth is closer to 300 Hz.

Where one has the option to adjust the bandwidth, I suspect using an
indicated bandwidth of 350+ Hz to obtain a true 1 dB bandwidth of at
least 300 Hz will provide the best overall performance for 75 baud.

73,

    ... Joe, W4TV


On 9/19/2011 5:04 PM, Peter Laws wrote:
> On Mon, Sep 19, 2011 at 12:52, Kok Chen<chen@mac.com>  wrote:
>
>> If the modem that is optimized for 45.45 baud does not change the internal 
>> filter for 75 baud operation, the SNR does not go up.  However, something 
>> even worse happens since a 75 baud signal will cause some pretty severe 
>> inter-symbol interference (a signal QRM'ing itself, so to speak :-).  You 
>> can even see this effect when using 45.45 baud, if you use a crystal filter 
>> in a rig that is too narrow, without also having an appropriate shape for 
>> binary 45.45 baud data.
>>
>
>
> I wondered about this when I was trying to copy K9MUG/4.  All I did
> was set MMTTY to "decode" 45.45 bauds.  I didn't fiddle with any of
> the other 10,000 settings on the config screen (for fear I'd delete
> half the internet if I did).
>
> I'm guessing that what I really need is a 75-baud "profile" in MMTTY
> that optimizes things for that speed (along the lines of 23-Hz shift
> profile that's distributed with the software).
>
> Anyone done that that is willing to share?
>
>
_______________________________________________
RTTY mailing list
RTTY@contesting.com
http://lists.contesting.com/mailman/listinfo/rtty

<Prev in Thread] Current Thread [Next in Thread>