Complex 16-bit radix-4 FFT

UM0304

14/19

4.3

Arguments and variables

The algorithm takes three values on input:

●

A pointer x to the input data consisting of N complex numbers x[0], ...., x[N-1]. Each

complex number is a 32-bit integer containing the 16-bit real part followed by the 16-bit

imaginary part.

●

A pointer y to the output buffer for the transformed array to be stored. This buffer is the

same in size as the input buffer x, with same order as in input buffer for the real part as

well as the imaginary part.

●

An integer nBin giving the base 2 logarithm of the number of points N in the Fourier

Transform.

4.4

Function

extern void cr4fft1k16_arm9e(void *pssOUT, void *pssIN, int nBin);

The input to the function is an array of complex data with each even index the real part and

imaginary part being in odd index, e.g., x[0] real part and it imaginary part x[1]; x[2] real part

and it imaginary part x[3]; and so on … Similarly, the output is also in the same order.

The block outline of this FFT algorithm is shown in Table 5.

4.4.1

Calling the FFT function from C

To call the FFT routine from C, include both 91x_dsp.c and 91x_dsp.h which declares the

type structure and the function FFT as:

extern void cr4fft1k16_arm9e(void *pssOUT, void *pssIN, int nBin);

Dual sine max neg.

Two sines at 800, 1200;

Fs =25600;

0x8001

4 bins of 0xE000 (+/-

0x2)

Dual sine max pos.

Two sines at 800, 1200;

Fs =25600;

0x7FFF

4 bins of 0x2000 (+/-

0x2)

Table 4.

Bit growth of FFT routine

Case name

Description

Input amplitude

Output bin amplitude

Table 5.

Block outline of FFT routine

Name

Function

Implementation

save_context

Save all registers used by the

routine

d_main

Main loop; transfer 16 bytes per

iteration

single loop

d_swit

Second level switch: transfer 4,

8 or 12 bytes

linear routine

d_switch

Third level switch: transfer 1, 2

or 3 bytes

switch case

restore_context

Restore all registers