Date : Mon, 12 Feb 1996 18:02:59 +0100
From : Ole Stauning <os@...>
Subject: Re: Decoding BBC tapes...
> From beeb-emulators-request@... Mon Feb 12 17:30:57 1996
> Resent-Date: Mon, 12 Feb 1996 16:26:32 GMT
> X-Sender: robert@...
> X-Mailer: Windows Eudora Pro Version 2.2 (32)
> Mime-Version: 1.0
> Content-Type> : > text/plain> ; > charset="us-ascii">
> On Mon, 12 Feb 1996, Ole Stauning wrote:
> >Maby it is possible to take smaller sound-samples, maby 2400 times
pr. sec. Every
> >sound-sample should be long enough to enclose two successive zero
crossings. You
> >only have to measure the time between the two crossings. Of course
this method
> >would be more sensitive to noise, but then you could try to take more
sound-samples
> >pr. sec. (make it a variable). I would love to program this, but i
am long away
> >from home and my BBC :-(
>
> I didn't get this. If you sample at 2400 Hz, you should be lucky to detect
> anything at all. The "0" frequency is exactly 2400 Hz, so your sample can
> (in theory) be all zeroes. If you're lucky, "0" will show as "++" (or
> "--"), while "1" will show as either "+-" or "-+".
Thats not what i meant. When i wrote sound-sample i meant several successive
samples, each sound-sample should of course be sampled with a freq. >
twice the
highest freq. that you expect on the tape. Like this
[----] [----] [----] [----]
|<------- T1 ------->| |<T2>|
T1 is the time between two successive sound-samples.
T2 is the duration of a sound-sample.
On Mon, 12 Feb 1996, Mark Cooke wrote:
> Nice - what you can do to avoid noise is to ignore zero crossings withing
> a certain time frame of the initial one:
>
> ,--------. .--------
> ----------' \ /
> `--------'
>
> ^ ^ ^
> 0 crossing
>
> XXXXXX XXXXXX XXXXXX
> Ignore regions.
>
> The ignore region should be (probably) just smaller than the shortest
> time between zero crossings that you expect to see. This should add noise
> resilience, as you ignore the times when the waveform is closest to 0.
Yes! First you find a zero crossing then you calculate where the next
zero crossing would be (exactly) using the shortest expected time. Then
from this
expected point you search in both directions some specified number of
samples.
If you do not find any crossing you can assume that it is the lower freq.
Regards
Ole
--
* legal notice: Microsoft Network is prohibited from redistributing this *
* work in any form, in whole or in part without a license. License to *
* distribute this work is available to Microsoft at $499. Transmission *
* without permission constitutes an agreement to these terms. *
--