[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Discuss-gnuradio] (no subject)
From: |
Eric Blossom |
Subject: |
Re: [Discuss-gnuradio] (no subject) |
Date: |
Thu, 1 Dec 2005 14:08:45 -0800 |
User-agent: |
Mutt/1.5.6i |
On Thu, Dec 01, 2005 at 01:53:15PM -0500, address@hidden wrote:
> On Wed, 30 Nov 2005 15:00:34 -0800, Eric Blossom wrote:
> > The calling sequence for the gaussian filter design changed.
> > Even if ggmsk-test were fixed to use it, it doesn't work anywhere
> > nearly as good as the new code.
>
> I would like to use the new code in gmsk2 but I can't implement the CRC32
> and the whiten/dewhiten in the microprocessor used in the transmitter. The
> original engineer on this project, who left the company and that I am
> replacing, selected a real poor microprocessor.
Then use CRC16. It only requires a constant 256 byte table.
The table required for whitening only needs to be as long as your
longest message. Shouldn't be a problem, even on a 8051.
What processor *did* he choose? ;)
> Do you think that I can use the same basic idea of the gmsk2 framer and
> sync whithout the use of the CRC32 and whiten?
Without whitening you may get no transitions in the data (e.g., all
zeros or all ones). This results in a single tone which has zero
information for recovering the clock. Whitening is cheap. I assume
you've got some flash to store code and/or data.
Eric