Prev: Incorrect eigenvector when using eigs with very large, sparse matrix
Next: result of sum(LargeArray) changes from execution to execution
From: Jan Simon on 9 Aug 2010 08:09 Dear Debora, > I have an emg signal that has a huge peak in a very low frequency, something between 0.5 and 1.2 hz. After filtering with a high pass butterworth of 20 hz and order 9, I could only reduce this peak, that is now half of its size, but still way higher then the signal. I've tried to change remove all signal under 5 Hz, but it did not work. Then, can anyone help me on how to remove a specific frequency on fft? Sorry for the stupid question: *All* EMGs I've worked with have a large peak at 0.5 to 1.2 Hz, because I'm examinating gait cycles which have these fequencies for fast and slow walking. Are you sure that such a big peak in your data is junk? Otherwise deleting everything but the expectations it is not a valid strategy get reliable scientific results. What does it mean, that "it did not work to remove all signal under 5 Hz"? What have you done exactly? Kind regards, Jan |