2.11.5 corr1
Menu Information
Analysis: Signal Processing: Correlation
Brief Information
Compute correlation of two signals
Command Line Usage
1. corr1 ix1:=Col(1) ix2:=Col(2);
2. corr1 ix1:=Col(1) ix2:=Col(2) type:=circular;
3. corr1 ix1:=Col(1) ix2:=Col(2) norm:=1;
4. corr1 ix1:=Col(1) ix2:=Col(2) type:=circular norm:=1 oy:=(col(3),col(4));
X-Function Execution Options
Please refer to the page for additional option switches when accessing the x-function from script
Variables
Display
Name
|
Variable
Name
|
I/O
and
Type
|
Default
Value
|
Description
|
| Input
|
ix1
|
Input
vector
|
<active>
|
Specifies the first input signal.
|
| Input
|
ix2
|
Input
vector
|
<active>
|
Specifies the second input signal. If you would like to calculate auto-correlation, this should be the same as Input Signal 1.
|
| Sampling Interval
|
interval
|
Input
double
|
1.0
|
Specifies the sampling interval. The default is <auto>, which corresponds to an automatically-computed interval.
|
| Type
|
type
|
Input
int
|
line
|
Specifies whether to compute a linear correlation or a circular correlation.
Option list
- linear :Linear
- This is suitable for signals whose data points outside the input range can be viewed as zeros.
- circular:Circular
- This is best suited for signals that repeat periodically.
|
| Normalize
|
norm
|
Input
int
|
0
|
Specifies whether the result should be normalized
|
| Output
|
oy
|
Output
XYRange
|
<new>
|
Specifies the output range
See the syntax here.
|
Examples
- To compute the auto-correlation of signals in Column 1 of the active worksheet, use the script command:
corr1 ix1:=col(1) ix2:=col(1)
- To change the method of correlation, open the dialog from the Origin menu, Analysis: Signal Processing: FFT: Correlation. In the dialog, select the options and click OK.
- To perform correlation to data using a pre-saved smoothing theme file, save your preferences in the corr1 dialog, and then execute it by typing the following script command, using your own saved-theme title:
corr1 -t "my correlation theme.oth"
// Use correlation to detect time delay between two signals
//Create a workbook and import the sample data
newbook sheet:=0 name:=Correlation;
newsheet xy:="YYYXY" labels:="Signal 1|Signal 2|Shifted Signal 2";
fname$ = System.path.program$ +"Samples\Signal Processing\Correlation.dat";
impasc;
string bkname$ = %H; // save the book name as plotting will create new window to change %H
//Calcualte correlation
corr1 ix1:=1 ix2:=2 type:=line oy:=(4,5);
//Get the maximum correlation and calculate the time difference
stats ix:=5 max:=maxcorr;
shift=col(4)[list(maxcorr, col(5))] + 1;
//Shift Signal 2 and plot
copydata irng:=col(2) orng:=col(3)[shift];
%a=nameof(col(3))$;
plotxy iy:=1 plot:=200 ogl:=Layer1;
plotxy iy:=[bkname$]1!col(3) plot:=200 ogl:=Layer1;
set %a -c 3;
More Information
For more information, please refer to our User Guide.
Related X-Functions
fft1, ifft1
Keywords:linear, circular
|