[fpc-pascal] Good FFT example anywhere?
Žilvinas Ledas
zilvinas.ledas at dict.lt
Sun Apr 9 10:22:00 CEST 2017
On 2017-04-09 11:06, Bo Berglund wrote:
> I am looking for some good example of FFT functions in pascal but it
> seems like what Google serves up is rather old and refers to
> Turbo-pascal and the like...
> So maybe someone here knows about some open-source example of FFT
> using FreePascal (or Delphi)?
Hi,
I was using attached unit some 7 years ago for various speech processing
purposes.
Best regards,
Žilvinas
>
> I want to analyze the frequency content of transient responses
> measured using a 24 bit A/D converter. It will produce 8192 samples
> for each measurement.
>
> Any suggestions welcome!
>
> PS: I have saved this list from Sept 2003 and I searched it for the
> word FFT without success, except the hits on the word offtopic.. DS
>
>
-------------- next part --------------
(*==========================================================================
fourier.pas - Don Cross <dcross at intersrv.com>
This is a Turbo Pascal Unit for calculating the Fast Fourier Transform
(FFT) and the Inverse Fast Fourier Transform (IFFT).
Visit the following URL for the latest version of this code.
This page also has a C/C++ version, and a brief discussion of the
theory behind the FFT algorithm.
http://www.intersrv.com/~dcross/fft.html#pascal
Revision history [most recent first]:
1996 December 11 [Don Cross]
Improved documentation of the procedure CalcFrequency.
Fixed some messed up comments in procedure ifft.
1996 December 6 [Don Cross]
Made procedure 'fft_integer' more efficient when buffer size changes
in successive calls: the buffer is now only resized when the input
has more samples, not a differing number of samples.
Also changed the way 'fft_integer_cleanup' works so that it is
more "bullet-proof".
1996 December 4 [Don Cross]
Adding the procedure 'CalcFrequency', which calculates the FFT
at a specific frequency index p=0..n-1, instead of the whole
FFT. This is O(n) instead of O(n*log(n)).
1996 November 30 [Don Cross]
Adding a routine to allow FFT of an input array of integers.
It is called 'fft_integer'.
1996 November 18 [Don Cross]
Added some comments.
1996 November 17 [Don Cross]
Wrote and debugged first version.
==========================================================================*)
//{$N+,E+} (* Allows code to use type 'double' and run on any iX86 machine *)
{$R-} (* Turn off range checking...we violate array bounds rules *)
unit Fourier;
interface
//uses definition;
(*---------------------------------------------------------------------------
procedure fft
Calculates the Fast Fourier Transform of the array of complex numbers
represented by 'RealIn' and 'ImagIn' to produce the output complex
numbers in 'RealOut' and 'ImagOut'.
---------------------------------------------------------------------------*)
procedure fft (
NumSamples: word; { must be a positive integer power of 2 }
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: array of double;
var ImagOut: array of double );
(*---------------------------------------------------------------------------
procedure ifft
Calculates the Inverse Fast Fourier Transform of the array of complex
numbers represented by 'RealIn' and 'ImagIn' to produce the output complex
numbers in 'RealOut' and 'ImagOut'.
---------------------------------------------------------------------------*)
procedure ifft (
NumSamples: word; { must be a positive integer power of 2 }
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: array of double;
var ImagOut: array of double );
(*---------------------------------------------------------------------------
procedure fft_integer
Same as procedure fft, but uses integer input arrays instead of
double. Make sure you call fft_integer_cleanup after the last
time you call fft_integer to free up memory it allocates.
---------------------------------------------------------------------------*)
procedure fft_integer (
NumSamples: word;
var RealIn: array of integer;
var ImagIn: array of integer;
var RealOut: array of double;
var ImagOut: array of double );
(*--------------------------------------------------------------------------
procedure fft_integer_cleanup
If you call the procedure 'fft_integer', you must call
'fft_integer_cleanup' after the last time you call 'fft_integer'
in order to free up dynamic memory.
--------------------------------------------------------------------------*)
procedure fft_integer_cleanup;
(*--------------------------------------------------------------------------
procedure CalcFrequency
This procedure calculates the complex frequency sample at a given
index directly. Use this instead of 'fft' when you only need one
or two frequency samples, not the whole spectrum.
It is also useful for calculating the Discrete Fourier Transform (DFT)
of a number of data which is not an integer power of 2. For example,
you could calculate the DFT of 100 points instead of rounding up to
128 and padding the extra 28 array slots with zeroes.
--------------------------------------------------------------------------*)
procedure CalcFrequency (
NumSamples: word; { can be any positive integer }
FrequencyIndex: word; { must be in the range 0 .. NumSamples-1 }
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: double;
var ImagOut: double );
implementation
function IsPowerOfTwo ( x: word ): boolean;
var i, y: word;
begin
y := 2;
for i := 1 to 15 do begin
if x = y then begin
IsPowerOfTwo := TRUE;
exit;
end;
y := y SHL 1;
end;
IsPowerOfTwo := FALSE;
end;
function NumberOfBitsNeeded ( PowerOfTwo: word ): word;
var i: word;
begin
Result:=1;
for i := 0 to 16 do begin
if (PowerOfTwo AND (1 SHL i)) <> 0 then begin
Result := i;
exit;
end;
end;
end;
function ReverseBits ( index, NumBits: word ): word;
var i, rev: word;
begin
rev := 0;
for i := 0 to NumBits-1 do begin
rev := (rev SHL 1) OR (index AND 1);
index := index SHR 1;
end;
ReverseBits := rev;
end;
procedure FourierTransform (
AngleNumerator: double;
NumSamples: word;
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: array of double;
var ImagOut: array of double );
var
NumBits, i, j, k, n, BlockSize, BlockEnd: word;
delta_angle, delta_ar: double;
alpha, beta: double;
tr, ti, ar, ai: double;
begin
if not IsPowerOfTwo(NumSamples) or (NumSamples<2) then begin
write ( 'Error in procedure Fourier: NumSamples=', NumSamples );
writeln ( ' is not a positive integer power of 2.' );
halt;
end;
NumBits := NumberOfBitsNeeded (NumSamples);
for i := 0 to NumSamples-1 do begin
j := ReverseBits ( i, NumBits );
RealOut[j] := RealIn[i];
ImagOut[j] := ImagIn[i];
end;
BlockEnd := 1;
BlockSize := 2;
while BlockSize <= NumSamples do begin
delta_angle := AngleNumerator / BlockSize;
alpha := sin ( 0.5 * delta_angle );
alpha := 2.0 * alpha * alpha;
beta := sin ( delta_angle );
i := 0;
while i < NumSamples do begin
ar := 1.0; (* cos(0) *)
ai := 0.0; (* sin(0) *)
j := i;
for n := 0 to BlockEnd-1 do begin
k := j + BlockEnd;
tr := ar*RealOut[k] - ai*ImagOut[k];
ti := ar*ImagOut[k] + ai*RealOut[k];
RealOut[k] := RealOut[j] - tr;
ImagOut[k] := ImagOut[j] - ti;
RealOut[j] := RealOut[j] + tr;
ImagOut[j] := ImagOut[j] + ti;
delta_ar := alpha*ar + beta*ai;
ai := ai - (alpha*ai - beta*ar);
ar := ar - delta_ar;
INC(j);
end;
i := i + BlockSize;
end;
BlockEnd := BlockSize;
BlockSize := BlockSize SHL 1;
end;
end;
procedure fft (
NumSamples: word;
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: array of double;
var ImagOut: array of double );
begin
FourierTransform ( 2*PI, NumSamples, RealIn, ImagIn, RealOut, ImagOut );
end;
procedure ifft (
NumSamples: word;
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: array of double;
var ImagOut: array of double );
var
i: word;
begin
FourierTransform ( -2*PI, NumSamples, RealIn, ImagIn, RealOut, ImagOut );
(* Normalize the resulting time samples... *)
for i := 0 to NumSamples-1 do begin
RealOut[i] := RealOut[i] / NumSamples;
ImagOut[i] := ImagOut[i] / NumSamples;
end;
end;
type
doubleArray = array [0..0] of double;
var
RealTemp, ImagTemp: ^doubleArray;
TempArraySize: word;
procedure fft_integer (
NumSamples: word;
var RealIn: array of integer;
var ImagIn: array of integer;
var RealOut: array of double;
var ImagOut: array of double );
var
i: word;
begin
if NumSamples > TempArraySize then begin
fft_integer_cleanup; { free up memory in case we already have some. }
GetMem ( RealTemp, NumSamples * sizeof(double) );
GetMem ( ImagTemp, NumSamples * sizeof(double) );
TempArraySize := NumSamples;
end;
for i := 0 to NumSamples-1 do begin
RealTemp^[i] := RealIn[i];
ImagTemp^[i] := ImagIn[i];
end;
FourierTransform (
2*PI,
NumSamples,
RealTemp^, ImagTemp^,
RealOut, ImagOut );
end;
procedure fft_integer_cleanup;
begin
if TempArraySize > 0 then begin
if RealTemp <> NIL then begin
FreeMem ( RealTemp, TempArraySize * sizeof(double) );
RealTemp := NIL;
end;
if ImagTemp <> NIL then begin
FreeMem ( ImagTemp, TempArraySize * sizeof(double) );
ImagTemp := NIL;
end;
TempArraySize := 0;
end;
end;
procedure CalcFrequency (
NumSamples: word; { must be integer power of 2 }
FrequencyIndex: word; { must be in the range 0 .. NumSamples-1 }
var RealIn: array of double;
var ImagIn: array of double;
var RealOut: double;
var ImagOut: double );
var
k: word;
cos1, cos2, cos3, theta, beta: double;
sin1, sin2, sin3: double;
begin
RealOut := 0.0;
ImagOut := 0.0;
theta := 2*PI * FrequencyIndex / NumSamples;
sin1 := sin ( -2 * theta );
sin2 := sin ( -theta );
cos1 := cos ( -2 * theta );
cos2 := cos ( -theta );
beta := 2 * cos2;
for k := 0 to NumSamples-1 do begin
{ Update trig values }
sin3 := beta*sin2 - sin1;
sin1 := sin2;
sin2 := sin3;
cos3 := beta*cos2 - cos1;
cos1 := cos2;
cos2 := cos3;
RealOut := RealOut + RealIn[k]*cos3 - ImagIn[k]*sin3;
ImagOut := ImagOut + ImagIn[k]*cos3 + RealIn[k]*sin3;
end;
end;
begin { Unit initialization code }
TempArraySize := 0; {flag that buffers RealTemp, RealImag not allocated}
RealTemp := NIL;
ImagTemp := NIL;
end.
(*--- end of file fourier.pas ---*)
More information about the fpc-pascal
mailing list