This was our third experiment in which first performed 4 point FFT and then 8 point FFT. The concept was made clear earlier by the butterfly diagrams taught in class. We also performed IFFT to verfy our original signal.
We calculated real and imaginary values separately -> A+iB and C+iD where A and C are real and B and D are imaginary parts for furhter multiplacations.
We compared the real and complex additions as well as the complex and real multiplications of FFT with DFT and came to a conclusion that FFT improves operational efficicency by parallel processing.
Codes can found at :-https://drive.google.com/open?id=0Bzfvoo_rjoa8S19TN2V1SE9kckk
FFT is faster because number of computations are reduced in it as compared to DFT.
ReplyDeleteFFT is useful for long signals, but fails for shorter ones, since it just increases the computations for smaller signals.
ReplyDeleteWe can perform 6 point or 12 point FFT as well . It is also called as composite FFT.
ReplyDeleteFFT RADIX can not necessary be powers of two
ReplyDeleteFFT is efficient algorithm for computing DFT.
ReplyDelete