{"id":2494,"date":"2009-11-13T13:46:00","date_gmt":"2009-11-13T13:46:00","guid":{"rendered":"https:\/\/blogs.mathworks.com\/pick\/2009\/11\/13\/this-peaks-my-interest\/"},"modified":"2009-11-13T13:46:00","modified_gmt":"2009-11-13T13:46:00","slug":"this-peaks-my-interest","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/pick\/2009\/11\/13\/this-peaks-my-interest\/","title":{"rendered":"This “Peaks” My Interest"},"content":{"rendered":"
\r\n

Jiro<\/a>'s pick this week is PeakFinder<\/tt><\/a> by Nate Yoder<\/a>.\r\n <\/p>\r\n

\"What? Another peak finder?\"<\/i> you might say. Some of you may classify this as one of those utilities that has been created by many people over the years,\r\n like sudoku<\/a> and waitbar<\/a>. Well, peak finding happens to be something dear to my heart.\r\n <\/p>\r\n

I have been using MATLAB for almost 10 years since my first year of graduate school. I initially learned by trying to decipher\r\n my advisor's code. One day, I was struggling to write some code for finding peaks in my data.\r\n <\/p>

% Sample data<\/span>\r\nt = 0:0.01:10;\r\nx = sin(2*t) - 3*cos(3.8*t);<\/pre>
That's when my advisor showed me his code:<\/p>
dx = diff(x);          % get differences between consecutive points<\/span>\r\npkIDX = (dx(1:end-1) >= 0) & (dx(2:end) < 0); % look for slope changes<\/span>\r\npkIDX = [dx(1)<0, pkIDX, dx(end)>=0];         % deal with edges<\/span>\r\nplot(t, x, t(pkIDX), x(pkIDX), 'ro'<\/span>);<\/pre> 
This was an eye-opener and was the moment I experienced the power of vector operation for the first time. The way I code in\r\n      MATLAB had changed from that point on. ... So when I see \"peak finding\", it brings back memories.\r\n   <\/p>\r\n   
There are quite a few File Exchange entries for finding peaks (and valleys), including two previous POTW selections: FPEAK<\/tt><\/a> and EXTREMA<\/tt><\/a>. But I really like peakfinder<\/tt> by Nate. Not only does his code deal with noisy data (my algorithm above will be useless if the signal is noisy), but also\r\n      his coding practice is quite solid. He has a great help section, robust error-checking of input arguments, and variable input\r\n      and output arguments for ease of use.\r\n   <\/p>
xNoise = x + 0.3*sin(40*t);   % add a few more bumps<\/span>\r\npeakfinder(xNoise);<\/pre> 
I looked through a few peak finding entries, but I'm sure I may have missed some. Feel free to let me know of others you really\r\n      like here<\/a>.\r\n   <\/p>