[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Packet Decoder Breaking when adding noise

From: Achilleas Anastasopoulos
Subject: Re: [Discuss-gnuradio] Packet Decoder Breaking when adding noise
Date: Tue, 20 Nov 2018 15:08:34 -0500

Hi Mathew,

the random source outputs bytes with numbers 0 or 1 (NOT 2).
This single bit information is "packed" into a byte that is the input to the QAM modulator.
The QAM modulator internally breaks up these bytes and uses every 4 bits to signal the 16QAM constellation.


On Tue, Nov 20, 2018 at 3:03 PM Matthew J. Brandsema <address@hidden> wrote:

When I switched AC coupling off, it just went blank.

So instead, I saved the waveforms to a file and plotted the result in Matlab.  See attached for plots.  The blue curve are the data before modulating, and the red x's are the data after demodulating.  They are lined up!

Thanks so much for the help in getting this working!!!  I noticed the random numbers are only 0, 1 and 2.  How the heck does this work for 16 QAM?  How does it encode 3 numbers to 16 symbols?

My next step is to transmit and receive using two Ettus radios next.  I got them to transmit and receive but I can't seem to find a delay that will make them sync up like I did with the file you sent. 

I saved their IQ waveforms to see how out of sync they are by doing a correlation and the correlation peak seem to be changing between data collects. Does anyone have any idea of what might be happening? (Sorry to add another question, I have been trying all day to get this working).

From: Achilleas Anastasopoulos [address@hidden]
Sent: Tuesday, November 20, 2018 1:25 PM
To: Matthew J. Brandsema; address@hidden
Subject: Re: [Discuss-gnuradio] Packet Decoder Breaking when adding noise

Can you switch off the AC-coupling, and check the scope plot again.
Also, maybe useful to have markers instead of just lines.

The two streams should be perfectly synchronized (for low noise) and as noise gets higher
some of the demodulated bits are in error. This is perfectly OK.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]