FFT algorithm implementation on an FPGA for processing 2^N points from the BINGO telescope.

First, you need to add vsim to your PATH. So, open the terminal and type this command:

sudo nano .bashrc

NOTE: nano it's just a simple terminal text editor.

Now that we are at bashrc file, you just have to include the vsim in the PATH list. For doing these, add the line bellow to the end of the file:

export PATH=/home/$USER/intelFPGA_lite/20.1/modelsim_ase/linuxaloem:$PATH

Write the changes with Ctrl+O and Enter, and then exit with Ctrl+X.

After adding to PATH, update the bashrc file with:

source .bashrc

As we can execute vsim now, we need to give permission to run_vsim.sh do his work. So, to do that, run the command bellow at this repo directory:

chmod +x run_vsim.sh

and that's it. run_vsim.sh can be executed by the python code now. Therefore, let's move on to the next settings.

If this is your first time running this project, you're going to have to build it. Start open the Quartus project at R22SDF/FFT.qpf, build the project with Ctrl+L, this can take several minutes.

After that, goes to Tools -> Run Simulation Tool -> RTL simulation to open the ModelSim and build the simulation for being used in the run_vsim.sh later.

Everything is built and you just have to do that again if you modify some R22SDF files, otherwise, you can test with different input from the python code.

The simulation is pretty forwards, after all those settings, you just have to run the python code and test with different inputs.

This is how it looks like:

But if you prefer, you can use ModelSim normally, like this picture below:

• Sistenix/FFT
• The Fast Fourier Transform (FFT)
• Recursive FFT in Python Convertible to Verilog/VHDL
• Namake FFT