discuss-gnuradio
[Top][All Lists]
Advanced

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

Re: USRP X410 Underflows and Overflows


From: Marcus Müller
Subject: Re: USRP X410 Underflows and Overflows
Date: Wed, 14 Feb 2024 14:00:33 +0100
User-agent: Mozilla Thunderbird

Hi Olivia,

generally, Python blocks for signal processing are pretty limited in speed; so I'd really recommend you compare your code's performance with Signal Source (sawtooth) -> frequency modulator or Signal Source (sawtooth) -> multiply with itself (giving you the square) -> phase modulator (maybe you remember Lars Gölz's chirp modulation Bachelor Thesis: that's what he did!).

Other than that, you're sadly not sharing the code of your chirp generation, nor the flow graph, but only partial screenshots of text – that's not very helpful to help you!

Best,
Marcus

On 14.02.24 12:48, Mammadova, Olivia wrote:
Hello together,

i wrote a code for a linear chirp and added the code to the python generated by grc, because GNU Radio has no building block for a linear chirp  except using Signal source and VCO Block.

I tried to simulate the linear chirp signal on USRP X410 and send  the chirp signa to USRP X410 l and check the output with the spectrum analyzer. Unfortunately, the simulation causes underflows and overflows for  the sample rate starting at 1 MHz and i cannot receive a proper linear chirp at UHD Source. Moreover, the CPU Usage during the python simulation is below 10%..  When i stop the simulation and want to restart the simulation, RFoncError occurs: Attempting to reconnect input port.

I tried to resize the buffer using sudo nano /etc/sysctl.d/99-custom.conf and added net.core.rmem_max=250000000 and applied the changes using sudo sysctl -p /etc/sysctl.d/99-custom.conf.  The the buffer size is resized, but the overflows and underflows are yet occuring and the cpu usage is below 10%.

I am sending you the screenshots of my simulation console and hope that you 
help me!

Thank you in advance!

Best regards,

Olivia Mammadova




reply via email to

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