How can I group positive peaks in groups by their similar values ?

So I have both positive and negative peaks for any given signal. I am interested in grouping them into positive and negative

pks_max =[

6.8285

6.0007

13.4501

13.4501

22.5547

23.3825

32.0733

31.6596

48.6273

49.4551

66.4228

85.4598

85.4598];

Here is my code for positive peaks, given that there is a same amount of positives as negatives, I could use the same code. This code worked when there were a same amount of values for each group, but it doesn't for my new case

% Get distance of every element to every other element.

distances_max = pdist2(pks_max,pks_max);

% Find out which pairs are within 1 of each other.

within2_max = distances_max > 0 & distances_max < 5;

% Erase upper triangle to get rid of redundancy

numElements_max = numel(pks_max);

j_max = logical(triu(ones(numElements_max, numElements_max), 0));

within2_max(j_max) = 0;

% Label each group with an ID number.

[labeledGroups_max, numGroups_max] = bwlabel(within2_max);

% Put each group into a cell array

for k = 1 : numGroups_max

[rows_max, columns_max] = find(labeledGroups_max == k);

indexes_max = unique([rows_max, columns_max]);

groups_max{k} = pks_max(indexes_max);

end

celldisp(groups_max); % Display the results in the command window.

[Image Analyst]

What is your definition of a negative peak? Is it a valley (a negative going peak), or is it a peak (upright) that has a peak value that happens to be negative?

[Heirleking]

I need to have different groups for numbers: 6s, 10s, 20s, 30s, 40s, 60s and 80s

[Image Analyst]

OK, that did not answer my question at all.

Let's try it another way.

pks_max =[
    6.8285
    6.0007
    13.4501
    13.4501
    22.5547
    23.3825
    32.0733
    31.6596
    48.6273
    49.4551
    66.4228
    85.4598
    85.4598];
plot(pks_max, 'b-');
grid on;

In the plot above, which are the positive peaks and which are the negative peaks?

[Heirleking]

I did not show the negative peaks it is a signal. All I need to know is how to group these array of numbers

[Image Analyst]

I see you accepted David's reply below so it appears that what you were really after was the histogram, you just didn't know the name of the function, and instead used words like peaks and groups. Well, glad you learned a new term and that David solved it for you. And thanks in advance for Accepting his answer to award him reputation points. Of course I suspect he has a prerelease version of the Mind Reading Toolbox.

Answer 1

[David Hill]

[~,~,bins]=histcounts(pks_max,[0 10 20 30 40 50 60 70 80 90]);%or whatever bin edges you want
u=unique(bins);
for k=1:numel(u)
    c{k}=pks_max(bins==u(k));%puts groups into different cell elements
end

[Heirleking]

Thanks this works!