Re: [Discuss-gnuradio] Correlation Estimator Block

[Cinaed Simson]

Suppose someone asked you to implement your flow graph on a transmitter
and receiver separated by an air gap.

Where would you tap into the signal with the Correlation Estimator?

I would tap into the signal at the end of the receiver - at the virtual
source with the decoded signal - or the very end of the flow chart.

And you've already demonstrated the output of the flow graph is the same
as input, i.e., you can reliably transit and receive a BPSK signal.

Instead of trying to cram all these block into 1 flow graph, there's an
easier way.

Replace the entire flow graph with a Channel Model. Use a complex vector
equal to the sum of the data and preamble.  Feed the complex vector into
a complex Stream to Tagged Stream block and then feed the tagged stream
into the Channel Model. Place the Correlation Estimator on the other
side of the Channel Model and you have something that begins to look like

     <gnuradio_root>/share/gnuradio/examples/digital/pack
/example_corr_est.grc

Use your Constellation Object to configure the Modulate Vector for BPSK.

And after you've work out all the details of the Correlation Estimator,
you can graph it back onto the original flow graph.

-- Cinaed


On 05/24/2017 10:32 AM, Mojtaba Mansour Abadi wrote:
> Hi Everyone,
> 
> 
> I am trying to benefit "Correlation Estimator" block to tag the preamble
> of the transmit bit sequence in a QPSK-based system
> 
> 
> The modulation is done as:
> Data + Preamble -> Constellation Modulator
> 
> 
> The way I am doing the demodulation is:
> AGC -> FLL Band-Edge -> Correlation Estimator -> Correlation Estimator
> -> Polyphase Clock Sync -> CMA Equaliser -> Costas Loop
> 
> The hard decoding is done as:
> Constellation Decoder -> Differential Decoder -> Map -> Unpacked to Packed
> 
> After I execute the flowgraph, for a few seconds everything works fine
> and I receive the correct sequences. However, after a while, the
> received bits are not correct and their pattern changes periodically.
> 
> When I bypass the "Correlation Estimator" block, everything works fine
> and I receive the correct data.
> 
> I am confused. Is the correlation block suppose to deteriorate the
> demodulation performance?
> 
> The flowgraph is attached to the email.
> 
> 
> -- 
> Regards,
> Mansour.
> 
> https://www.linkedin.com/in/mojtaba-mansour-abadi-4311b451
> 
> 
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>