Hi everyone, Can anybody help me to plot ??

The data I am using has 21 columns, the 1st column is Time in seconds. The sampling rate is 0.33 sec. I have data for four hours. The rest 20 columns are data values corresponding to the Time. I want to plot each of the 20 columns corresponding to the time in one figure to see the trend in them, if any. How do I do it ?? I also want to give color to the data for each hour.

Answer 1

[Geoff Hayes]

Adarsh - do you really want to give colour for each hour? Wouldn't you rather have colour for each of the 20 columns of data and see how they vary over time?

You can try the following to load your data and plot it

 myData = xlsread('pbri_Oct1-20.xlsx');
 plot(myData(:,1),myData(:,2:end),'.');

In the above, we read the data from the Excel file and then plot each of the 20 columns versus time (column one) in one call to plot. Each column is assigned a different colour and you can a trend for each in the plot.

You may instead want to create four separate plots for each hour of data.

[Add]

hey hayes, thanks for the help. but it plots all columns at exact same position and each one overlaps on the other. To have a clear view of image I want to plot them using vertical offset . can you help me further to get a output like this ??

[Geoff Hayes]

You could try the following and plot each column individually and just add an offset to each.

 figure;
 hold on;
 for k=2:size(myData,2)
     plot(myData(:,1),myData(:,k)+2*k,'.');
 end

You can then adjust the 2*k offset as you see fit.

If you wish to change the colour of the plot given a different hour you could use another for loop to plot by hour...if you know where each new hour begins. For simplicity, using your sample file, I will just divide the input by four and assume that each block is size(myData,1)/4 rows in length.

 myData = xlsread('pbri_Oct1-20.xlsx');
 figure;
 hold on;
 blockSize = floor(size(myData,1)/4);
 for b=1:4
     blockColour = rand(1,3);
     for k=2:size(myData,2)
         plot(myData(1+blockSize*(b-1):blockSize*b,1),...
              myData(1+blockSize*(b-1):blockSize*b,k)+2*k,...
              '.',...
              'Color', blockColour);
     end
 end

You can adjust the above to suit your needs. (The colours can be hard-coded instead of relying on random colours; the block size should be replaced with the actual position in your data when the new hour begins; etc.)

[Add]

Thanks mate. I appreciate your work. Here, as you have introduced breaks in colour at the 1/4th size of the block, how can I give the colour to it as time in the 1st column changes from one whole digit to the next, i.e.. 18 to 19, 19 to 20....and so on. Simply, How to introduce breaks acccordingly to a change in time ??

[Geoff Hayes]

Since the first column corresponds to time, then use the floor function to round down each real number to the nearest integer that is less than or equal to that value. i.e.

 timeValues = floor(myData(:,1));

Now timeValues is an array of integers and you are interested in whenever there is a change. So use diff to determine a difference from one integer to another

 timeChangesIdcs = find(diff(timeValues)~=0) + 1;

Using find we are able to get the those indices of where a change in time occurs. (A one is added because the array returned from diff has one less element than timeValues due to the nature of that function.)