Error with find command for a matrix

Hello,

I have data points from a DAQ with a time and signal. My goal is to find the period of my data points. To do this, I wanted to find the maximum peak value and the minimum peak value. If I have these, I should be able to find the corresponding time to those values, which would be half my period since the shape of my data is basically a repeating sine wave.

Here is what I have so far:

_____________________________________________________

clc;

format short

%Compose matrix with time vecotr and signal vector

Test1Signal1=[Test1_Time, Test1_Signal1];

%Find the maximum value

a=max(Test1Signal1,[],1);

a=a(:,2)

%Find the mimumum value

b=min(Test1Signal1,[],1);

b=b(:,2)

%Find the time that corresponds to the maximum value (a)

[r c]=find(Test1Signal1,a)

%Find the time that corresponds to the minimum value (b)

[r c]=find(Test1Signal1,b)

_____________________________________________________

The problem is, I keep getting an error. The error I'm getting is as follows:

??? Error using ==> find Second argument must be a positive scalar integer.

Error in ==> Test1 at 16 [r c]=find(Test1Signal1,a)

I cannot figure out why the code is giving me this error. The value of a (the maximum) is a positive number. Again, I want this to give me the row and column that corresponds to the maximum value so that I can find the time from the original matrix that matches this value.

If that can get figured out, I might have another problem. The minimum value is a negative number. I'm not sure if the find command is going to work on a negative number, but that might be another problem once I get there. Any help would be greatly appreciated.

Thanks

Answer 1

[Paulo Silva]

[r c]=find(Test1Signal1==a)

instead of

[r c]=find(Test1Signal1,a)

Answer 2

[Walter Roberson]

find() does not take two arguments that are an array to search in and the value to search for. The first argument to find() is interpreted as a logical condition with all non-zero values corresponding to "true". Usually find() then returns the locations of all of the positions that are considered true, but find() takes an optional numeric argument that must be a positive integer that tells it how many indices to return. Thus, find(Test1Signal1,a) is requesting that find() return the first "a" locations of Test1Signal that are non-zero. For example, find(Test1Signal1,2) would return the indices of two non-zeros.

The code you are looking for is

[r c] = find(Test1Signal1==a)

I suggest, though, that you consider

[a, minloc] = min(Test1_Signal1)
[b, maxloc] = max(Test1_Signal1)

and then the times would be Test1_Time(minloc) and Test1_Time(maxloc)

Your logic is fragile, though: due to noise or due to the sampling rate not being exactly a multiple of your frequencies, the minimum over the entire signal is not necessarily in the same period as the maximum value.