Sps in receiver

Hi all guys,

this time :), my question is regarding soem reference book or any other
kind of literature, where I can find about how digital receiver should
be
designed regarding sps. Namely, my current receiver is more or less
simulation of some analog circuitry that I understand better than DSP.
For
example, after demodulation I use a block that is apprximation of diode
min
max circuit for setting the right threshold for getting dgital signal
out
of “analog” demodulator output. After that my bit slicer works more or
less
like the one found in the UART. For this reasons, the signal needs to be
highly oversampled, for example, symbol rate is 19.2kb and I use
sampling
frequency of 500ksps. When lowering sampling frequency to 250ksps, the
performance of the receiver degrades a lot, plus probably to many
processing effort than it should be.

Thanx,

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Nemanja,

uff, that’s quite a lot of text I hardly understand.
Could you please confirm:

  • You want to design a receiver
  • the receiver should have an Analog-to-digital converter
  • You already have a design for mixing from RF to baseband

The rest I don’t understand. What is your question? How is it related
to GNU Radio?
But it really does sound exciting!

Greetings,
Marcus

On 06.12.2013 11:26, Nemanja S. wrote:

like the one found in the UART. For this reasons, the signal needs
mailing list [email protected]
Discuss-gnuradio Info Page

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.15 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJSobNZAAoJEAFxB7BbsDrLpsMH/2btq+dIEaFNalI4zfvLWOvV
bJQn/aEcH/R+cPHmAUbj3zm9ro5aATCWSyW3nUPjPyaXvtJDFnTHx1lCnH8GzDGK
Ab2KzcwOJnPZTCnrQZmeiWlHsPMe2QnAw2omsuHdqGoDaom/z3vSOB3Vt33pB2/4
MnQPB51nzsdlVMXtFoi3y5gUC/WclSz2U60TUnOBfA6hbuZI/mdbY4oOC7FtvqMf
YogYlcTQpuKgZThV5q0wVJWpBlkOscK3Cg/xcjLkYtXEC6LaQvN2iokNB3jiMcWY
c+9IZ2IuN5OlUwMvrA4gTv+Mf9Loo/xlJoFGrp0oMwQeabdJ+JDHewvgr6Fl6q8=
=HNQz
-----END PGP SIGNATURE-----

Ok, i should put more info. I have USRP and WBX., and I designed
receiver
for operating withing 434 MHz.
Since I am more hardware oriented electrical engineer, I didn’t know how
to
make it in “correct” way, and there are a few blocks there
that are based on simulating hardware that I know how functionate. So, I
use sampling frequency of 500k, and after demodulator, I have a block
that
works literally as diode min max circuit and is usually implemented in
some
receivers. That block, extracts threshold (which is half way between min
and max). After that, the same input is compared to threshold signal in
order to derive signal that has only zero level and one level (so
digital
signal). Derived “digital” signal is sent to something that I call clock
syncronizer, and there I search for the edges in order to allign my
sampling period, and I do sampling in order to get symbol values.
So my question is what would be the correct way to design this, because
I
think that my approach is definately not correct one and in my opinion
too
robust.

Yes, that’s true, but actually I am using USRP/WBX and GNURADIO,
everything
works, but I think I can make it better if I redesign it in the spirit
of
SDR.

I want to receive both fsk and on-off, and at the moment I have two
parallel processing paths.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ah ok, I think I can make out some misunderstandings:

  • You use an USRP / WBX System
  • You want to use GNU Radio to decode a signal
    • transmitted at 434 MHz
    • having a symbol rate of 19.2 kHz
    • being binary coded
    • that binary coding being on-off keying

Is that correct?

On 06.12.2013 12:49, Nemanja S. wrote:

and one level (so digital signal). Derived “digital” signal is sent
Hi Nemanja,

after demodulation I use a block that is apprximation of
Thanx,
Discuss-gnuradio Info Page

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.15 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJSobsEAAoJEAFxB7BbsDrLUxMH/3qZqxND7Q0TYtAh6Ztc5pg8
ns+D3C3swusKoGAdR+ajXrRjSX2rr0fHcEmGP9dmW+WUc0FohY/abHMpSO28IZld
Xrj9saBxv00IlG27HJqmZcLJQCnm4k80cZoKwkiLE0Z6XGFfVRhNFdvU4lbLk6Rs
nWEgi7aKozY1L6kJgJjTtx48o6R2SJWWkolPP5eZemblcvrOnSb60DN+lRkWFfmA
ZNW8n7ph0pqblAaWEGg8umfnj7TEGGjyjy0JBZGDRclpaHBMNAngQ3W3tOxeAD3t
EhaqLwLAPvXh1QTs10K91DFdrYiIDQZNsXl6LgSo5skJa2R0LD1yHrrGgC4flcg=
=O5vA
-----END PGP SIGNATURE-----

Hope you (Marcus) didn’t change your mind not help me :slight_smile:
Actually I just wanted to ask for some literature, and the question is
not
directly connected to GNU radio, but rather indirectly.

Best