Re: [Discuss-gnuradio] UHD source + sink PROBLEM

[Marcus D. Leech]

On 01/10/2012 05:48 PM, UB wrote:
> Here is my code for transmitting data from a file:
> from gnuradio import gr
> from gnuradio import uhd
>
> symbolrate=0.5*10**5
> samplerate=20*symbolrate  # 1M
> fg=gr.top_block()
> src=gr.file_source(gr.sizeof_gr_complex*1,'data.txt',False)
> snk=uhd.usrp_sink('serial=1R24X9U1',uhd.io_type.COMPLEX_FLOAT32,1)
> snk.set_samp_rate(samplerate)
> snk.set_center_freq(2.4*10**9)
> fg.connect(src,snk)
> fg.run()
>
> And I run the uhd_fft.py on another USRP.. i constantly get a peak at centr
> frequency for data being sent, why?
> http://old.nabble.com/file/p33115067/data.txt data.txt
Many different debugging approaches:

a) Plot your data with an FFT sink--rather than a usrp_sink. Is there a 
spectral peak at DC?

b) There will always be a small amount of DC offset in both directions 
(RX and TX) that will
    cause an anomaly around DC.  You can null this out to a certain 
extent using the calibrate
    routines (assuming you have a USRP2/N2XX and either a WBX or SBX card).


c) You can always offset tune your TX, to take the DC offset anomaly 
away from your main
    transmit bandwidth.  Similarly in RX.


--
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org