Hi Colin, The B2xx/AD936x has built in DC offset calibration that may be causing this, and generally you will want to either offset tune so your signal is not at the exact center frequency, but still within your passband, or use the LO offset functionality.
Using the LO offset method is the easiest, by adding the argument "lo_offset=xxx" to your device string, such as: ./rx_samples_to_file --args "type=b200,master_clock_rate=50e6,lo_offset=25e6" .... Regards, Nate Temple On Tue, Mar 27, 2018 at 9:02 AM, de Vrieze, Colin via USRP-users < usrp-users@lists.ettus.com> wrote: > Hi all, > > > When using the rx_samples_to_file example script, I'm experiencing weird > discontinuities in the envelope of the signal, which I cannot get rid of. > > > My intention is to visualize the CFO between transmitter and receiver. TX > transmits a random BPSK-modulated at 5.75 GHz. I am using a lab-grade > vector signal generator, and sampling the transmitted signal with a > spectrum analyzer shows that it is clean. > > > On the receiver side, I use a USRP B200 as RX, connected via USB3. I run > the rx_samples_to_file script for 100ms with srate=50MHz. I use port RF A > RX2. I chose the buffer size to match the number of total samples per > recording to exclude buffering issues. I double checked with print > statements in the receive script that the whole recording fits in the > buffer. After receiving all the data, I read the samples from the hard > disk, square and low-pass filter the values to extract the envelope signal. > > > In any case, the magnitude of the squared signal should be constant. In > case of no CFO, the I and Q signal should both be constant, too. In case of > CFO, the I and Q signals should be sinusoids with a frequency equal to > twice the CFO. Again, sampling with a spectrum analyzer shows just that. > > > However, the signal recorded by the B200 shows strange discontinuities. > They tend to have a somewhat regular structure, but the shape changes. The > stream indicates no under-/overflows. > > > I get the same effect with different sample rates, too. For example, I > tried srate=master_clock_rate=40e6 or srate=0.5*master_clock_rate=50e6. I > already chose a super long set-up time to make sure the CFO has settled and > I printed out the metadata of recv_stream to make sure there are no issues. > The same effect happens if I only record for 10ms. Please find the output > of the rx_samples_to_file script below. > > > Has anyone seen this before or does anyone have an idea how to solve it? > > > Cheers, > > Colin > > > --- > > > Creating the usrp device with: serial=EAR04Z7B2,master_clock_rate=50e6... > Detected Device: B200 > Operating over USB 3. > Initialize CODEC control... > Initialize Radio control... > Performing register loopback test... > pass > Performing CODEC loopback test... > CODEC loopback test passed > Asking for clock rate 50.000000 MHz... > Actually got clock rate 50.000000 MHz. > Performing timer loopback test... > Timer loopback test passed > Using Device: Single USRP: > Device: B-Series Device > Mboard 0: B200 > RX Channel: 0 > RX DSP: 0 > RX Dboard: A > RX Subdev: FE-RX2 > TX Channel: 0 > TX DSP: 0 > TX Dboard: A > TX Subdev: FE-TX2 > Setting RX Rate: 50.000000 Msps... > Actual RX Rate: 50.000000 Msps... > Setting RX Freq: 5750.000000 MHz... > Actual RX Freq: 5750.000000 MHz... > Setting RX Gain: 55.000000 dB... > Actual RX Gain: 55.000000 dB... > Waiting for "lo_locked": ++++++++++++++++++++++++++++++++++++++++++++++++++ > locked. > Number of requested samples: 525000 > Buffer size: 525000 > Tick > Number of received samples: 525000 > Has timespec: Yes Time of first sample: 10.2594 > Fragmented: No Fragmentation offset: 0 > Start of burst: No End of burst: No > Error Code: ERROR_CODE_NONE Out of sequence: No > Done! > > > > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > >
_______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
