LimeSuiteNG on Ubuntu2025_04

Hello
I installed LimeSuiteNG on a fresh Ubuntu2025_04
from scratch with these steps:

  1. Install gedit (Optional)
    sudo apt install gedit
  2. Install Git
    sudo apt install git
  3. Install C, C++ Compiler and environment
    sudo apt-get install build-essential
  4. Install cmake
    sudo apt install cmake
  5. Install make
    sudo apt install make
    Always the newest version!
  6. Install USBLib
    sudo apt install libusb-1.0-0-dev
  7. Install GnuPlot
    sudo apt install gnuplot
  8. Install wxWidgets
    sudo apt install libwxgtk3.2-dev
  9. Install LimeSuiteNG
    git clone GitHub - myriadrf/LimeSuiteNG
    cd LimeSiteNG
    sudo ./install_dependencies.sh
    cmake -B build
    cd build
    make
    sudo make install
    sudo ldconfig

Now I can start LimeGUI and use the Module fftViewer.

But in Step9, I get errors and warnings:
sudo ./install_dependencies.sh
E: Unable to locate package linux-headers-generic-hwe-25.04
E: Couldn’t find any package by glob ‘linux-headers-generic-hwe-25.04’
E: Couldn’t find any package by regex ‘linux-headers-generic-hwe-25.04’

make
/home/khk/LimeSuiteNG/src/chips/LMS7002M/LMS7002M.cpp:222:17: warning: ‘uint16_t lime::GetWritableBits(uint16_t)’ defined but not used [-Wunused-function]
222 | static uint16_t GetWritableBits(uint16_t addr)
| ^~~~~~~~~~~~~~~

Has everybody an idea what it is with the messages
in step 9. or can I ignore that?

Thanks

linux-kernel-headers package is required for PCIe driver compilation. If you aren’t going to use LimeSDR PCIe devices you can ignore it.
Otherwise you can try installing “sudo apt-get install linux-kernel-headers” package. Ubuntu 25.04 might have different name for the package.

Hello Ricardas, thanks for your note.
I have two more questions about LimeSuiteNG

a) Is there a more detailed decription of the LimeSuiteNT?
I have found only a more globally function of that.

b) On the module fftviewer, I have seen that the
FFT Amplitude scale is dBFS with the steps
-12, -23, -34, -46, -58, -69, -88, -92, -104
Is this correct, because the step width is different?!?
Futhermore, when I feed a calibrated signal to the input and
compare the calculated value with the FFT-Value
I have a difference of some(±5dB)dB.

Thanks

PS: The next week, I am not inhouse.

What kind of description you need? API, code examples, use cases?

it’s correct, that’s just the numbers of the graph grid, they change if you zoom in or move the graph view.

FFTviewer shows dBFS values (of the digital signal), it is not equal to absolute analog RF signal dBm power, and the dBFS values you see are affected by the RF chip gain settings. So the values of the signal generator and the FFT viewer are not the same.