Hi Pedro,
I don't use the Python interface so I am not experienced on it.DC leakage can
happen on both Tx side and Rx side, that is, in your current setup, it can be
in USRP or in the spectrum analyzer.The spike in your spectrum is about 20dBm
higher than the signal, which is pretty significant. Normally, DC is not so
big. Maybe you have put extra power at DC?Can you plot the spectrum of the
original baseband signal?
Marcus is the expert to answer USRP questions. Marcus, could you please shed
any light on this?
Regards,Hongwei
On Wednesday, 28 February 2024 at 01:14:38 GMT, Pedro Vieira (Gmail -
Geral) <pav.vie...@gmail.com> wrote:
Dear Hongwei, and Ron:
Thank you for the comments and suggestions.
I've been working on some configurations, but still with the presence of the
spark/spike. It also seems to me that it could be something related to the
generation of the OFDM signal..
Any suggestion?
Attach the code I use and the signal file, in case you have the opportunity to
check.
The master clock rate is 200 Msps and the sampling rate is 25 Msps
Some of the features of the equipment and programs I am using are:
- I'm using the Spyder IDE version 5.5.1
- UHD library version: 4.6.0.0
usrp.get_usrp_tx_info()
Oct[134]:
{'mboard_id': 'X310',
'mboard_name': '',
'mboard_serial': 'F61798',
'module_serial': 'F61798',
'tx_antenna': 'TX/RX',
'tx_id': 'CBX (0x0066)',
'tx_ref_power_key': 'x3xx_pwr_generic_tx_tx+rx',
'tx_ref_power_serial': 'F5B3E5#0',
'tx_serial': 'F5B3E5',
'tx_subdev_name': 'CBX TX',
'tx_subdev_spec': 'A:0 B:0'}
Kind regards and thank you in advance,
Pedro A.
De: zhou <hwz...@yahoo.com>
Enviada em: sábado, 24 de fevereiro de 2024 09:09
Para: usrp-users@lists.ettus.com; Pedro Vieira (Gmail - Geral)
<pav.vie...@gmail.com>
Assunto: Re: [USRP-users] RES: OFDM signal transmission by x310 presents a peak
Hi Pedro,
Interestingly, just found Marcus answered a similar question a few years back:
https://dsp.stackexchange.com/questions/30562/large-spike-at-the-center-frequency-when-using-ettus-x310
Your calibration range is pretty small. I normally calibrate from 500M - 6G
such that it can support multiple applications. It will take some time, but
calibrating 1G should be quick.
USRP may not be able to generate LO exactly at 3.5GHz. In such case, it will
try to generate LO as close as possible.
Which UHD version are you using?
What are the master clock rate and sampling rate?
If calibration doesn't have any effect on your received signal, the USRP device
may be faulty, or the analyzer is problematic.
A few things you can try:
1. send 0s and you may see just a spike at the centre without signal;
2. allocate your modulation symbols to one side, lower or upper, of the OFDM
resource grid, the signal spectrum will go to one side and should not be
impacted by the central spike;
3. Use USRP as receiver. Loop back Tx to Rx of the same USRP. Plot the spectrum
of the received signal to see if there is central spike.
Kind regards,
Hongwei
On Saturday, 24 February 2024 at 00:59:34 GMT, Pedro Vieira (Gmail - Geral)
<pav.vie...@gmail.com> wrote:
Thank you for the quick response.
Based on the information received, it was possible to begin to understand what
was happening:
1) Proceed with self-calibration:
uhd_cal_tx_dc_offset --verbose --freq_start=3.45e+9 --freq_stop=3.55e+9
--freq_step=1e+6 --subdev=A:0
uhd_cal_tx_dc_offset --verbose --freq_start=3.45e+9 --freq_stop=3.55e+9
--freq_step=1e+6 --subdev=B:0
Calibration files were generated: tx_dc_F5B3E5.cal and tx_dc_F59F87.cal
2) In the uhd library I found the command uhd.types.TuneRequest. And in the
gnuradio library the command is the suggested example: uhd.tune_request().
However, I didn't notice any changes (see below)
Any ideas what I might be overlooking? Thank you in advance.
De: zhou <hwz...@yahoo.com>
Enviada em: sexta-feira, 23 de fevereiro de 2024 17:45
Para: usrp-users@lists.ettus.com; Pedro Vieira <pav.vie...@gmail.com>
Assunto: Re: [USRP-users] OFDM signal transmission by x310 presents a peak
Hi Pedro,
It could be DC leakage. Try to run the calibration commands. For X310, you need
to loopback connect Tx and Rx antennas with 30dB attenuators.
Regards,
Hongwei
On Friday, 23 February 2024 at 01:52:37 GMT, Pedro Vieira
<pav.vie...@gmail.com> wrote:
An ofdm signal, which is generated in python, presents a peak in the central
part of the spectrum. What could it be?
This same behavior appears on USRP X310 and HackRFOne.
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
