[Top][All Lists]

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

Re: [Discuss-gnuradio] Receiving signal from a signal generator

From: John Orlando
Subject: Re: [Discuss-gnuradio] Receiving signal from a signal generator
Date: Fri, 17 Feb 2012 10:44:11 -0600

On Fri, Feb 17, 2012 at 10:22 AM, Tom Rondeau <address@hidden> wrote:
On Fri, Feb 17, 2012 at 2:52 AM, guelord ingala <address@hidden> wrote:
I've got 2 units of USRP1 and DBSRX dboards. The two USRP are synchronized for master and slave. the uhd_fft.py is responding correctly.
Now I want to use a signal generator to provide 900 MHz. signal level is -20 dBm. From the gnuradio-companion, I created a flow graph using UHD-USRP Source connected to WX-GUI-FFT Sink. The setup are as follow:
 For USRP source:
        ID: uhd_usrp_source_0
        Device Addr: fpga=usrp1_fpga_4rx.rbf
        Sync: Don'y sink
        Clock rate: default
        Num Mboards:1
        Mbo Clock source: default
        Mbo Time source:default
        Mbo Subdevice Spec:
        Num Channel: 1
        Sample rate (sps): 32K
        Cho Center freq: 900M
        Cho Gain:30 dB
        Cho Antenna: RXA
        Cho Bandwidth:10M

For FFT Sink:
       Sample rate:32k
       Baseband freq:900M

With this setup, only noise is plotted in the FFT. I don't see the signal at 900MHz. And even if I remove the daughterboard from the usrp, the result is unchanged: Noise only.
Can you please tell me what is wrong with the setup, and what to do if I want to use the side B. Also which USRP must I use: the master, the slave or both. Please help.

I think your sample rate is way too low for the USRP. When you run the program, you should see a warning being printed that you tried to set the sample rate to 32 kHz but the actual sample rate was something else. This could cause some confusion in your setup.

Also, make sure that you are going into the right antenna port on the USRP.

Finally, a -20 dBm signal and 30 dB of gain in the USRP is a really loud signal. You might be swamping your front end. I'd put the input signal down to -50 dBm to start with and lower the gain. Up the gain slowly as necessary.

One other potential issue here: if you're tuning the LO to 900 MHz, and injecting a tone at exactly 900 MHz from a signal generator, the signal downconverted signal would end up sitting exactly at DC, and may be removed by DC offset correction if it is enabled.  Once you've widened up your sample rate a bit per Tom's suggestion (try something on the order of a couple of MHz, or whatever decimation ratio would get you to something in this range), try offsetting your input tone to 900.1 MHz and see if it shows up.


John Orlando
CEO/System Architect
Epiq Solutions

reply via email to

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