Interpolate and synchronize two time series

Hello,

I'm trying to match two different timeseries to one common time-vector. The problem is, there are two matrices which I want to match. First of all I'll give an example:

matrix_A
TimeA DataA_1 DataA_2 DataA_3
0.27  63      57      12.3
0.31  65      47      11.3
0.54  69      53      11.7
0.82  71      51      11.9
0.98  70      52      12.1
1.24  69      49      12.4
--------------------------
matrix_B
TimeB DataB_1 DataB_2
0.13  113     23.7
0.54  112     25.8
0.69  110     24.4
0.85  115     23.5
1.03  113     25.9

Now i want to match this two series of data to one matrix with equal time gaps, for example 0.25.

My idea is: First interpolate both data-series to get values at 0; 0.25; 0.5; 0.75; 1.0; 1.25 ... and then safe them in one single matrix, that looks like:

matrix_finish
Time DataA_1 DataA_2 DataA_3 DataB_1 DataB_2
0    ...     ...     ...     ...     ...
0.25
0.5
0.75
1.0
1.25
...

I know, there are functions in matlab like synchronize and interp1 or something, but it looks like I am to stupid to use them correctly for my problem.

Thanks for any help.

Answer 1

[jonas]

1. Create two timetables using table2timetable (you have to converted the time to duration or datetime format)
2. Use synchronize OR retime with a common time vector.

Easy peasy. Upload some data and I will show you how.

[Lukas Huber]

Good morning jonas,

thanks for your fast answer. Here are two examples of my data as txt-files. At the moment I safe the files manually as Excel-Files (Dataserie1.xlsx and Dataserie2.xlsx) and then I scan them in with the MATLAB-function xlsread, for example [matrixA,~,~] = xlsread('Dataserie1.xlsx'). This is necessary to switch the decimal-comma ',' to a decimal-point '.'.

Then I get two matrices with different time-stamps which I want to synchronize now and safe in one single matrix with equal time gaps of 0,250 sec as I explained in my question.

Would be nice if you show me how I have to use this timetable-function.

Thanks in advance!

[jonas]

Morning! Have a look at this and see if it looks alright. I parsed the text file for you directly in matlab.

% Read files
str1 = fileread('data1.txt');
str2 = fileread('data2.txt');

% Change decimal sign
str1 = regexprep(str1,',','.');
str2 = regexprep(str2,',','.');

%parse files
data1 = textscan(str1,'%{hh:mm:ss.SSS}T%f%f%f%f','headerlines',1,'delimiter',{'\t',' '},'multipledelimsasone',1);
data2 = textscan(str2,'%{hh:mm:ss.SSS}T%f%f%f','headerlines',1,'delimiter',{'\t',' '},'multipledelimsasone',1);

% Create timetables
TT1 = timetable(data1{1},data1{2:end});
TT2 = timetable(data2{1},data2{2:end});

% Common time vector
tc = (hours(8)+minutes(57)):milliseconds(250):(hours(8)+minutes(58))';
tc.Format = 'hh:mm:ss.SSS'

%synchronize using linear interpolation 
TT3 = synchronize(TT1,TT2,tc,'linear')

You may want to change the interpolation method.

[Lukas Huber]

Hey!

Thanks for your help, now it is working.

[jonas]

Great! My pleasure!