Mark Butcher (@mjbcswitzerland)
A couple of notes:1) There is an ARM /mbed forum but there are virtually no responses there (especially not from ARM)2) I have seen other people trying to do the...
The CMSIS code selects the FFT to be used based on the length as follows: switch (L) { case 16: case 128: case 1024: ...
HiCMSIS is ARM's "Cortex Microcontroller Software Interface Standard"https://developer.arm.com/embedded/cmsisI tried with different bit-reversal codes:static int...
Hi All
The ARM CMSIS FIXED POINT FFTs uses a bit reversal table at the end of the FFT calculation to transpose the output bins to their linear ordering....
Hi AllI decided not to use the RFFT and stick with the CFFT one - just constructing the required complex data at the input.This requires the additional copy and...
TimI think that you have hit the nail on the head. arm_cfft_f32() looks to really need complex input and doesn't just give complex output. Reading the documents...
MikeI am using single-precision floating point (the routine used is for that). Since I have an FPU in the final processor this is the best choice in my case rather...
RickMany thanks for the very wise advice. I have just performed the 4 tests.1.I get 360 rather than 512 but only DC.2.This test is as expected.3.This starts off...
Hi MikeThe magnitude calculation is arm_cmplx_mag_f32(fft_input_buffer, fft_magnitude_buffer, 256);, however it is only doing a root-mean-square of the imaginary...
Use this form to contact mjbcswitzerland
Before you can contact a member of the *Related Sites:
- You must be logged in (register here)
- You must confirm you email address
