nrf_dsp_cfft_radix8by2_f32()

void nrf_dsp_cfft_radix8by2_f32	(	const nrf_dsp_cfft_f32_t *	p_cfft,
		flt32_t *	p_input )

#include <include/nrf_edgeai/dsp/transform/fft/nrf_dsp_fft_f32.h>

The floating-point complex FFT uses a mixed-radix algorithm. Multiple radix-2 stages are performed along with a radix-1 stage, as needed. The algorithm supports lengths of [16, 128, 1024] and each length uses a different twiddle factor table.

Parameters

[in]	p_cfft	Pointer to instance of complex FFT structure nrf_dsp_cfft_f32_t
[in,out]	p_input	Pointer to buffer with input samples, the FFT computes in-place and will modify input buffer with FFT results. The results will be written to input array in the interleaved fashion: {real[0], imag[0], real[1], imag[1], ...}