Hello, i need some advice on how to condition input signal to calculate CFFT.
I have a signal, which is electrocardiograph signal with added sine wave. data type of an input signal is a float, so first i find it's maximum value, then divide the signal with that value from 1.01 (to have a range of 0.99996948 to -1.0. ) Then i convert it to fract16 format, perform fft, then
discard some unwanted spectral components which are related to noise, then multiply the result by 1024 (it's my fft length) or shift by 10.
After that i calculate inverse fft, and lastly convert to the output result to float. But the signal looks not like its supposed to.
i think there is a problem with overflow. I also attach .doc file which contains some comments and images of my signal. Can you tell me what am i doing wrong? I also attach my code in word file.
Rest regards, Andrius