The purpose of this application is to use ARM-cmsis-DSP on non-ARM processors.
To use the examples of this repository, it is necessary to go to the desired folder and run the make command
cd arm_matrix_example
make build
make run
In the gcc_fir_gnuplot
folder, there is a Python notebook file that simulate the FIR filter by used to cmsis_dsp wrapper and the C++ code to use the cmsis_dsp library is available.
We need to install Anaconda for Python execution library or use google colab for run python notebook.
And to run the C++ program, we need to install GNU Plot. for install gnuplot on debian machine, please run this commnad
sudo apt install gnuplot
finally, when execute command make build
and make run
on gcc_fir_gnuplot
folder, so show
this image:
it is to similar scipy FIR and cmsis-dsp wrapper output in the notebook file
finally we can check bot mean and variance on jupyter notebook and c++
for python notebook:
sin noisy mean: 0.0
sin noisy var: 0.51948
fir mean: 0.010776961394674657
fir var: 0.465838732679487
for C++
Mean SIN: -0.00129333
Variance SIN: 0.520287
Mean FIR: 0.0116191
Variance FIR: 0.460044