Home    Search    Reference Manuals    Return   

Users Manual for Program FTRANS

Table of Contents

Introduction

Program FTRANS is a general program for translating variables between time- and frequency-domain, in both Fourier series and Fourier transform. Program FTRANS can also filter a variable in frequency-domain in a user defined function "filter". The function "filter" contains the transfer function in the frequency-domain, and shall be written in complex format. Input data controlling the function in the filter is given in the input data command HIN.

Internal Calculation Procedure

Program FTRANS must not pass all steps below, the user can choose from which item FTRANS shall start and stop the calculation. The selection of starting point is made in the command TYPE_INFIL. The selection of stop point is made automatically, depending on results requested by the user. Request of results are made in the commands INTPFI, FOURFI, FSERFI, FILTFI and RESFIL,

1. Read input data file *.ftransf.

2. Read input variable from file INFIL.

3. Interpolate the input variable with an equidistant step, defined in input data parameter DX. Type of interpolation is defined by the user in command TYPE_INTERP.

4. Scale the variable with the scale factors SKALAX and SKALAY.

5. Shift the input variable vertically and horizontally with XSTART and YSTART.

6. Write interpolated and scaled input variable to file INTPFI.

7. Transform the input variable into the frequency-domain.

8. Write the Fourier transform to file FOURFI. The file is written according to FORMF and WRITE_FOUR.

9. Write the Fourier series to file FSERFI. The file is written according to FORMF and WRITE_FOUR.

10. Filter the Fourier transform in a complex filter given in HFUNC. Type of filtering is defined in command TYPE_FILT.

11. Write the filtered Fourier transform to file FILTFI. The file is written according to FORMF and WRITE_FOUR.

12. Transform the Fourier spectra back to time-domain again.

13. Write the filtered time-domain variable to file RESFIL. The file is written according to FORMF.

Input Data Commands

Input data are read in free format, valid separators between the input values are <space>, <comma>, <tab>, <equal sign> or <carriage return>. The commands can be written both in lower and upper case letters. The operation of the program is controlled by the commands described below; some of the commands also need arguments.

Summary of all main commands

DX

Equidistant step between the interpolated points in the X-axis. An equidistant step between two consecutive points in input data are necessary before making the FOURIER transformation, otherwise the FFT-algorithm will fail.
Input data command DX is only valid if TYPE_INFIL has been set equal to TIME_HIST.
If DX=0 or TYPE_INFIL= TIME_HIST_INTPL, DX will be set equal to the first distance X(2)-X(1).
In input data command TYPE_INTERP the user can select the type of interpolation to be used.

Declared= Real*4    Default= 0.

EOF

Command which terminates further input data reading.

IDENT1, IDENT2, IDENT3

Definition of ident lines.

Declared= Character*80    Default= Blank

INFIL

File containing the input data input variable.
The file is read at 2) above.
File consisting of value-pair complex or real. The file is read according to the format given in FORMIN. Lines starting with the #-sign are considered as being commentary lines.

Declared= Character*80    Default= Blank

FORMIN

Format in which file INFIL shall be read.
The string should be enclosed in parenthesis e.g. '(E15.3,20X,E15.3)'. The format specification shall be written in a way that it will fit into the FORTRAN read statement. The format specification given in the example above makes program FTRANS to read the first 15 characters in file INFIL into the X-variable, the next 20 characters will be skipped, the next 15 characters will be read into the Y-variable.
If TYPE_INFIL= TIME_HIST or TYPE_INFIL= TIME_HIST_INTPL it is possible to read INFIL in free format, if FORMIN is given one of the following values:

'(A,A)'            = reads column 1 & 2
'(A,X,A)'          = reads column 1 & 3
'(A,X,X,A)'        = reads column 1 & 4
'(A,X,X,X,A)'      = reads column 1 & 5
'(A,X,X,X,X,A)'    = reads column 1 & 6
'(A,X,X,X,X,X,A)'  = reads column 1 & 7
'(A,X,X,X,X,X,X,A)'= reads column 1 & 8

If TYPE_INFIL=FOUR* or FILT* FORMIN can be given the value CMPLX or SNGL. FORMIN= CMPLX implies that column 1 & 2 is read as a complex X-variable and column 3 & 4 is read as a complex Y-variable. FORMIN= SNGL implies that column 1 is read as the imaginary part of the complex X-variable and column 2 is read as the real part of the complex Y-variable. Maybe it seems unlogical to let the first column be the imaginary part of the X-variable, but the angular frequency in frequency domain has normally the real part equal to zero.

Declared= Character*80    Default= 'CMPLX'

INTPFI

File for writing the interpolated and scaled input variable to.
The output is written at 6) above.
No output is written if INTPFI is left blank.

Declared= Character*80    Default= Blank

FOURFI

File for writing the Fourier transform to.
The output is written at 8) above.
The Fourier transform is the integral from 0(zero) to T without multiplying with 1/T. The Fourier transform is calculated in complex precision, and the X-axis is expressed in [rad/s]. The user can control the form of the output of the spectra in the input data parameter WRITE_FOUR.
No output is written if FOURFI is left blank.

Declared= Character*80    Default= Blank

FSERFI

File for writing the Fourier series to.
The output is written at 8) above.
In the Fourier series is the amplitudes equal to the amplitudes in the time-domain signal, the levels of the amplitudes are as if they were put together by sine-waves. The Fourier series is calculated in complex precision, the real numbers corresponds to the amplitudes of the cosinus waves and the imaginary numbers corresponds to the amplitudes of the sinus waves. The X-axis is expressed in [rad/s]. The user can control the form of the output of the spectra in the input data parameter WRITE_FOUR.
No output is written if FSERFI is left blank.

Declared= Character*80    Default= Blank

FILTFI

File for writing the filtered Fourier transform to.
The output is written at 11) above.
No output is written if FILTFI is left blank.

Declared= Character*80    Default= Blank

NOTE

Commentary line.

Declared= Character*80    Default= Blank

RESFIL

File for writing the filtered output variable time-domain into.
The output is written at 13) above.
No output is written if RESFIL is left blank.

Declared= Character*80    Default= Blank

FORMF

Format in which the files INTPFI, FOURFI, FSERFI, FILTFI and RESFIL shall be written in.
The string should be written in FORTRAN syntax and shall be enclosed in parenthesis. All output data files are written in complex precision.

Declared= Character*80    Default= '(1p,2e18.10,5x,2e18.10)'

XSTART

Shift the X-axle of the input variable.

Declared= Real*4    Default= 0.

SKALAX

Scale the X-axis of the input variable.
The scale factor is used in the following way: Xcurve= (Xread-XSTART) * SKALAX Where: Xread are the values from from INFIL and Xcurve are the values used in further processing.

Declared= Real*4    Default= 1.

IEXP

Extrapolation of the input variable, in order to avoid end distortions.
Instead of using a Hanning window it is possible to extend the input variable read from INFIL in order to attenuate the signal in the start and in the end. This method does not remove any information about the input variable in its ends. IEXP are the number of points which the original input variable shall be extended with in further processing, but the output to the result files does only contain the original input variable, the extra points are only used in the calculations.

Declared= Real*4    Default= 0.

YSTART

Shift the Y-axle of the input variable.

Declared= Real*4    Default= 0.

SKALAY

Scale the Y-axis of the input variable.
The scale factor is used in the following way: Ycurve= (Yread-YSTART) * SKALAY Where: Yread are the values from from INFIL and Ycurve are the values used in further processing.

Declared= Real*4    Default= 1.

TYPE_FILT

Select type of filtering in 10) above.
TYPE_FILT can be given the following values:

FOUR	=	The complex transfer function in the filter is used as it is.
PSD	=	The complex transfer function in the filter is multiplied by its complex conjugate before filtering.

Declared= Character*6    Default= FOUR

TYPE_INTERP

Select type of interpolation in 3) above.
TYPE_INTERP can be given the following values:

LINEAR	=	Linear interpolation.
SPLINE	=	Spline interpolation.

Declared= Character*6    Default= LINEAR

TYPE_INFIL

Select starting point in the program.
TYPE_INFIL can be given the following values:

TIME_HIST	=	Start the program at 1) above. Begin with reading a time-history variable from file INFIL.
TIME_HIST_INTPL	=	Start the program at 4) above. Begin with reading an interpolated (equidistant) time-history variable from file INFIL.
FOUR_TRANS	=	Start the program at 10) above. Begin with reading a Fourier transformed variable from file INFIL.
FOUR_SERIE	=	Start the program at 10) above. Begin with reading a Fourier series from file INFIL.
FILT_FOUR_TRANS	=	Start the program at 12) above. Begin with reading a Fourier transformed variable from file INFIL.
FILT_FOUR_SERIE	=	Start the program at 12) above. Begin with reading a Fourier series from file INFIL.

Declared= Character*16    Default= TIME_HIST

WRITE_FOUR

Defines how to write the spectra on the output file.
WRITE_FOUR can be given the following values:

ROUND	=	Writes the spectra from 0 [rad/s] up to max. positive angular frequency, the spectra then continues from max. negative angular frequency up to 0 again.
STRAIGHT	=	Writes the spectra from max. negative angular frequency up to max. positive angular frequency via 0(zero).
POSITIVE	=	Writes only the positive frequencies from 0(zero) up to max. positive angular frequency.

Max. positive angular frequency is +π/DX and max. negative angular frequency is -π/DX. Where DX is the equidistant steps in the time-domain signal.

Declared= Character*16    Default= ROUND

HFUNC

Transfer function declared in FORTRAN as "COMPLEX FUNCTION".
The user can compile an own filter and link it into the FTRANS program, or he can use the predefined filters stored in file filter.f. Input data for the different filters is given via the vector HIN.

Declared= Character*80    Default= Blank

In file filter.f of FTRANS the following filters are available:

BUTTERWORTH_HIGH	Butterworth high pass filter
BUTTERWORTH_LOW	Butterworth low pass filter
DERIV	Derivation of input variable
FTRAPEZOID	Create a filter with trapezoidal shape
INTEG	Integration of input variable
HFILT	High pass filter
LFILT	Low pass filter
KLIPP	An ideal infinitely steep band-pass filter
IKLIPP	An ideal infinitely steep band-inhibiting filter
ISOL	Theoretical filter according to ISO 2631 for lateral accelerations
ISOV	Theoretical filter according to ISO 2631 for vertical accelerations
ISOL5	Filter according to ISO 2631,draft 5, lateral accelerations
ISOV5	Filter according to ISO 2631,draft 5, vertical accelerations
LSTRIX	Filter for generating the lateral errors due to designed track alignment
IMAUZ	Restores track alignment from a MAUZIN recording car
IPLASSER	Restores track alignment from Plasser&Theurer or Matisa

BUTTERWORTH_HIGH

Butterworth high pass filter.
HIN(1)= The order of the filter.
HIN(2)= Cut-off frequency [Hz]

BUTTERWORTH_LOW

Butterworth low pass filter.
HIN(1)= The order of the filter.
HIN(2)= Cut-off frequency [Hz]

DERIV

Derivation of input variable.
HIN= Not applicable

FTRAPEZOID

Create a filter with trapezoidal shape.
The transfer function of the filter is equal to 1.0 in the frequency range from 0 up to omega1 and from omega4 to infinity. In the frequency range from omega1 to omega2 the absolute value of the transfer function changes linearly from 1.0 to ampl2. In the frequency range from omega2 to omega3 the absolute value of the transfer function changes linearly from ampl2 to ampl3. In the frequency range from omega3 to omega4 the absolute value of the transfer function changes linearly from ampl3 back to 1.0. Input data is read in the following order:
HIN(1)= omega1
HIN(2)= omega2
HIN(3)= omega3
HIN(4)= omega4
HIN(5)= ampl2
HIN(6)= ampl3
The figure below shows an example how an unwanted spike in a frequency spectra can be removed. The black curve is the original frequency spectra, the blue curve is shows the spectra after filtering and the red line shows the inverse value of the transfer function of the filter.

INTEG

Integration of input variable.
HIN= Not applicable

HFILT

High pass filter.
HIN(1) = Number of poles.
HIN(2,3)= Pol #1 in complex precision. [rad/s]
HIN(4,5)= Pol #2 in complex precision. [rad/s]
HIN(6,7)= Pol #3 in complex precision. [rad/s]
HIN(8,9)= Pol #4 etc.

LFILT

Low pass filter.
HIN(1) = Number of poles.
HIN(2,3)= Pol #1 in complex precision. [rad/s]
HIN(4,5)= Pol #2 in complex precision. [rad/s]
HIN(6,7)= Pol #3 in complex precision. [rad/s]
HIN(8,9)= Pol #4 etc.

KLIPP

An ideal infinitely steep band-pass filter.
HIN(1)= Lower angular frequency [rad/s]
HIN(2)= Upper angular frequency [rad/s]

IKLIPP

An ideal infinitely steep band-inhibiting filter.
HIN(1)= Lower angular frequency [rad/s]
HIN(2)= Upper angular frequency [rad/s]

ISOL

Theoretical filter according to ISO 2631 for lateral accelerations, broad-band analysis.
HIN= Not applicable

ISOV

Theoretical filter according to ISO 2631 for vertical accelerations, broad-band analysis.
HIN= Not applicable

ISOL5

Filter according to ISO 2631 for lateral accelerations, draft 5.
HIN= Not applicable

ISOV5

Filter according to ISO 2631 for vertical accelerations, draft 5.
HIN= Not applicable

LSTRIX

Filter for generating the lateral errors due to designed track alignment, in the FIR-filter of the STRIX recording car.
HIN= Not applicable

IMAUZ

Restores the track alignment from measurements made by the MAUZIN recording car.
The transfer function of the MAUZIN recording car is zero on wave lengths which is a multiple of the axle distance 5[m]. The restoring of the original track alignment is therefore impossible, because the backward transfer function will go to infinity when the wave lengths is a multiple of 5[m]. In the HIN data vector, the user has the possibility to control the backward transfer function:

HIN(1)	=	K0	=	Max. inverted factor where the transfer function shall be cut off.
HIN(2)	=	K1	=	Inverted factor which shall be used when the transfer function exceeds 1/K0.
HIN(3)	=	WLONG	=	Limit for long waves. Waves longer than WLONG are attenuated.
HIN(4)	=	KLONG	=	Max. inverted amplifying factor for waves between LCAR and WLONG.
HIN(5)	=	WSHORT	=	Limit for short waves. Waves shorter than WSHORT will be amplified by the factor 1/KSHORT.
HIN(6)	=	KSHORT	=	Inverted amplifying factor for waves shorter than WSHORT.
HIN(7)	=	LCAR	=	Distance between axle #1 and #8. The distance is used for the definition of long waves, where the factor 1/KLONG should be used.
HIN(8)	=	SIGN	=	Overall amplifying factor. Can also be used for changing the sign of the MAUZIN registration.
HIN(9)	=	AWHEL0	=	Distance to the measuring axle. The distance shall be given relative to a reference point on the measuring car. The reference point can be any point on the measuring car, but it must be the common reference point for all axles.
HIN(10)	=	NWHELS	=	Number of counter axles to the measuring axle.
HIN(11:*)	=	AWHEL	=	Longitudinal distances to all counter axles.

Below follows a number of input data examples for filter IMAUZ. These recommendations refers to registrations digitalized every 1.[m] along the track. If the track is digitalized more dense than 1.[m], HIN(5) may be shorter than 3[m].
MAUZIN car, lateral irregularities, A-end leading:

HIN(1) = .5  (1/times)
HIN(2) = .5  (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 10. (m)
HIN(8) =  1. (units)
HIN(9) =  0. (m)
HIN(10)=  2  (st)
HIN(11)=  5. (m)
HIN(12)= -5. (m)

MAUZIN car, lateral irregularities, B-end leading:

HIN(1) = .5  (1/times)
HIN(2) = .5  (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 10. (m)
HIN(8) = -1. (units)
HIN(9) =  0. (m)
HIN(10)=  2  (st)
HIN(11)=  5. (m)
HIN(12)= -5. (m)

MAUZIN car, vertical irregularities, A-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 13.4(m)
HIN(8) = -1. (units)
HIN(9) =  0.675 (m)
HIN(10)=  8
HIN(11)=  6.710 (m)
HIN(12)=  4.210 (m)
HIN(13)=  2.075 (m)
HIN(14)=  0.675 (m)
HIN(15)= -0.675 (m)
HIN(16)= -2.075 (m)
HIN(17)= -4.210 (m)
HIN(18)= -6.710 (m)

MAUZIN car, vertical irregularities, B-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 13.4(m)
HIN(8) = -1. (units)
HIN(9) = -0.675 (m)
HIN(10)=  8
HIN(11)=  6.710 (m)
HIN(12)=  4.210 (m)
HIN(13)=  2.075 (m)
HIN(14)=  0.675 (m)
HIN(15)= -0.675 (m)
HIN(16)= -2.075 (m)
HIN(17)= -4.210 (m)
HIN(18)= -6.710 (m)

MAUZIN car, cant irregularities, A-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) =  .1 (1/times)
HIN(5) =  3  (m)
HIN(6) =  1  (1/times)
HIN(7) = 13.4   (m)
HIN(8) =  1.375 (units)
HIN(9) =  2.075 (m)
HIN(10)=  1
HIN(11)= -0.675 (m)

MAUZIN car, cant irregularities, B-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) =  .1 (1/times)
HIN(5) =  3  (m)
HIN(6) =  1  (1/times)
HIN(7) = 13.4   (m)
HIN(8) =  1.375 (units)
HIN(9) =  0.675 (m)
HIN(10)=  1
HIN(11)= -2.075 (m)

Plasser&Theurer car, cant irregularities, A-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3  (m)
HIN(6) =  1 (1/times)
HIN(7) = 10.  (m)
HIN(8) = -1.
HIN(9) = -3.5 (m)
HIN(10)=  1
HIN(11)=  0.  (m)

Plasser&Theurer car, cant irregularities, B-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 10.  (m)
HIN(8) = -1.
HIN(9) =  0.  (m)
HIN(10)=  1
HIN(11)=  3.5 (m)

IPLASSER

Restores the lateral and vertical track alignment from measurements made by a Plasser&Theurer or Matisa recording car.
The transfer function of the Plasser&Theurer recording car is much better compared to MAUZIN thanks to the asymmetric location of the measuring axle. In the HIN data vector, the user has the possibility to control the backward transfer function:

HIN(1)	=	K0	=	Max. inverted factor where the transfer function shall be cut off.
HIN(2)	=	K1	=	Inverted factor which shall be used when the transfer function exceeds 1/K0.
HIN(3)	=	WLONG	=	Limit for long waves. Waves longer than WLONG are attenuated.
HIN(4)	=	KLONG	=	Max. inverted amplifying factor for waves between LCAR and WLONG.
HIN(5)	=	WSHORT	=	Limit for short waves. Waves shorter than WSHORT will be amplified by the factor 1/KSHORT.
HIN(6)	=	KSHORT	=	Inverted amplifying factor for waves shorter than WSHORT.
HIN(7)	=	LCAR	=	Distance between first and last axle. The distance is used for the definition of long waves, where the factor 1/KLONG should be used.
HIN(8)	=	SIGN	=	Overall amplifying factor. Can also be used for changing the sign of the MAUZIN registration.
HIN(9)	=	AWHEL0	=	Distance to the measuring axle. The distance shall be given relative to a reference point on the measuring car. The reference point can be any point on the measuring car, but it must be the common reference point for all axles.
HIN(10)	=	NWHELS	=	Number of counter axles to the measuring axle. In filter IPLASSER NWHELS must be equal to 2.
HIN(11:12)	=	AWHEL	=	Longitudinal distances to the counter axles.

Below follows a number of input data examples for filter IPLASSER. These recommendations refers to registrations digitalized every 1.[m] along the track. If the track is digitalized more dense than 1.[m], HIN(5) may be shorter than 3[m].
Plasser&Theurer car, lateral irregularities, A-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1  (1/times)
HIN(7) = 20. (m)
HIN(8) =  1. (units)
HIN(9) =  0. (m)
HIN(10)=  2  (st)
HIN(11)= -10.(m)
HIN(12)=  10.(m)

Plasser&Theurer car, lateral irregularities, B-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 20. (m)
HIN(8) = -1. (units)
HIN(9) =  0. (m)
HIN(10)=  2  (st)
HIN(11)=-10. (m)
HIN(12)= 10. (m)

Plasser&Theurer car, vertical irregularities, A-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 10. (m)
HIN(8) = -1. (units)
HIN(9) = -3.5   (m)
HIN(10)=  2
HIN(11)= -5.    (m)
HIN(12)=  0.    (m)

Plasser&Theurer car, vertical irregularities, B-end leading:

HIN(1) = .25 (1/times)
HIN(2) = .25 (1/times)
HIN(3) = 400 (m) or track length/3.
HIN(4) = .04 (1/times)
HIN(5) =  3. (m)
HIN(6) =  1. (1/times)
HIN(7) = 10. (m)
HIN(8) = -1. (units)
HIN(9) =  3.5 (m)
HIN(10)=  2
HIN(11)=  5.  (m)
HIN(12)=  0.  (m)

HIN

Data for the subroutine containing the transfer function.
The different components in vector HIN has different meanings in different filters, see the documentation under HFUNC.

Declared= Real*4(1000)    Default= 1000*0.

Example:

Following example: Master.ftransf can be used as a master file:

##
##      Input data for program FTRANS
##

 IDENT1= Low-pass filtering of a variable in program FTRANS

##
##  Input reading phase
## - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
##
 INFIL=   ucat/icos_l5d2.cmplx          # Input data file
 FORMIN= '(A,A)'                        # Format when reading input data file
 FORMF = '(1P2E15.6,5X,2E15.6)'         # Format when writing output data file
 SKALAX= 1.  XSTART= 0.                 # Scale and shift the X-axle
 SKALAY= 1.  YSTART= 0.                 # Scale and shift the Y-axle
 DX  = 0.2                              # Equidistant steps in X-axle
 IEXP= 0                                # Elongation of the input variable

 TYPE_INFIL= TIME_HIST                  # Type of input data file
#TYPE_INFIL= TIME_HIST_INTPL
#TYPE_INFIL= FOUR_TRANS
#TYPE_INFIL= FILT_FOUR_TRANS

##
##  Interpolation phase
## - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
##
 TYPE_INTERP= LINEAR
 INTPFI= ucat/icos_l5d2.cmplx

##
##  Fourier transformation and Fourier series phase
## - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
##
 FOURFI= ucat/fcos_l5d2.cmplx
 FSERFI= ucat/Acos_l5d2.cmplx

##
##  Filtering phase
## - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
##
 TYPE_FILT= FOUR
#
 HFUNC= LFILT     HIN= 1, -1.25663706,                # low pass filter 1:st order
#
#HFUNC= LFILT     HIN= 2, -1.25663706, -1.25663706,   # low pass filter 2:d order
#                         -1.25663706, +1.25663706,
#
#HFUNC= HFILT     HIN= 1, -1.25663706,                # high pass filter 1:st order
#
#HFUNC= HFILT     HIN= 2, +1.25663706, +1.25663706,   # high pass filter 2:d order
#                         +1.25663706, -1.25663706,
#
 FILTFI= ucat/Fcos_l5d2.cmplx

##
##  Output of filtered time history
## - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
##
 RESFIL= ucat/rcos_l5d2.cmplx

 EOF