how to remove unwanted data from an arry

I have measured data that has some information outside the zone of interest.

the "wings" at either end of the plot attached are the unwanted data.

I am trying to write a script to analyze many data sets with similar issues is there a way to remove that from all data sets. I need to remove the same points from both the x and y sets of data.

[Andrew Shum]

Do you need to remove it from the arrays or just not plot them? Also, what type of criteria are you using to determine the values. Is it a continuous index range or specific, potentially isolated indices?

[Zachariah Johnson]

I would like to remove it from the array because I will later be processing the data from these arrays.

The values that I do not need are outside the maxima, <-125 and <125, but these maxima may move depending on how the data was collected.

sorry, this is from a different account I realized I was logged in on my old work account and would receive notifications

Answer 1

[Star Strider]

If you have the Signal Processing Toolbox, use the findpeaks function to identify the indices of the peaks. Then delete the data less than the first index and greater than the second index.

EDIT — In the absence of findpeaks, try this:

x = linspace(-140, 140);                    % Simulate Data
y = x.^2/1500;                              % Simulate Data
y(1:9) = y(10)-y(1:9);                      % Simulate Data
y(end-10:end) = y(end)-y(end-10:end);       % Simulate Data

L2 = fix(length(x)/2);                      % Split Data In Half
ixrng1 = 1:L2;                              % First Index Range
ixrng2 = L2:length(x);                      % Second Index Range

[~,ix1] = max(y(ixrng1));                   % First Maximum
[~,ix2] = max(y(ixrng2));                   % Second Maximum
ix2 = ix2+L2-1;                             % Correct ‘ix2’ For Offset

ClipWings = ix1:ix2;                        % Data Index Range With ‘Wings’ Deleted

figure(1)
plot(x, y)
hold on
plot(x(ix1), y(ix1), 'pg')
plot(x(ix2), y(ix2), 'pg')

figure(2)
plot(x(ClipWings), y(ClipWings))

Substitute your independent and dependent variables for my ‘x’ and ‘y’ respectively. The rest of my code should work without modification.

[Zachariah Johnson]

Star, thank you for the detailed description of the steps, I don't see where ix1 is defined?

[Star Strider]

My pleasure.

The start index ‘ix1’, is defined in this max call:

[~,ix1] = max(y(ixrng1));                   % First Maximum

[Zachariah Johnson]

ok I see, Thanks again for your help this worked great.

[Star Strider]

My pleasure.

If my Answer helped you solve your problem, please Accept it!

Answer 2

[Andrew Shum]

If you don't have the toolbox, you could try this.

idx1=find(xdata(1:round(numel(xdata)/2))<-125);
idx2=find(xdata(round(numel(xdata)/2):end)>125);
xdata=xdata((idx1(end)+1):(idx2(1)-1));
ydata=ydata((idx1(end)+1):(idx2(1)-1));

Edit 1:

Since it isn't at fixed values of x. It will have to be done slightly differently. The following should include the two maxima. If you don't want to include them, change the indices by 1.

idx1=find(ydata(1:round(numel(ydata)/2))==max(ydata(1:round(numel(xdata)/2))));
idx2=find(ydata(round(numel(ydata)/2):end)==max(ydata(round(numel(ydata)/2):end)));
xdata=xdata(idx1(end):idx2(1));
ydata=ydata(idx1(end):idx2(1));

Edit 2:

As you pointed out in the comments, the above solutions accidentally split the data in two.

splitidx=round(numel(ydata)/2);
idx1=find(ydata(1:splitidx)==max(ydata(1:splitidx)));
idx2=find(ydata(splitidx:end)==max(ydata(splitidx:end)));
idx2=idx2-1+splitidx;
xdata=xdata(idx1(end):idx2(1));
ydata=ydata(idx1(end):idx2(1));

[Zachariah Johnson]

Thank you, is there a way to find the turning points so I don't have to guess?

-125 and +125 was a guess

[Andrew Shum]

Sorry, I made a mistake in my initial reply. Please see the edited version.

[Andrew Shum]

Hang on.

[Andrew Shum]

Okay. The above should work based on what you have told us.

[Zachariah Johnson]

I'll give it a shot. Thank you, Andrew

[Zachariah Johnson]

Andrew, correction from my last post, it has cut my data set in half but other than that it worked. to remove the wings

[Andrew Shum]

My bad. If you still need a solution for future use, try the modified version I am about to add.