It as an alternative method for implementation of FIR Filter Design.
The desired frequency response Hd(w) is sampled at discrete values of frequency to obtain H(k). A factor of 2*pi has to be
added every time a sign change occurs in the magnitude of H(k).
The final output sequence is obtained by Inverse Discrete Fourier Transform (IDFT).
Hd(w)----------------H(k)------------------h(n).
(By sampling) (By IDFT)
The frequency sampling realization of filter is computationally more efficient mainly because the sample values are symmetric about k = N/2.
it would be more efficient if simulation results will also attached
ReplyDeleteThe frequency sampling realization of filter is computationally more efficient than the direct form realization as frequency sampling method eliminates the the co-efficient of H[k] which are zero.
ReplyDeletewell written.
ReplyDeleteThe algorithm for FSM is well-explained.
ReplyDeletealising may occur if nyquist criterion is not followed
ReplyDeleteFrequency sampling technique is one of the easiest methods for designing a FIR filter moreover it is computationally faster.
ReplyDeleteFSM is used in the cases where desired response in known.
ReplyDeleteIt is computationally more efficient because coefficients of H[k] becomes zero for w greater than cut off frequency. Thus only non - zero values are to be taken into consideration while obtaining h(n)
ReplyDeleteDFT is used in FSM which involves computation at fixed values of frequency. Hence, it is preferred.
ReplyDeleteEasy and preferred method for designing FIR filter
ReplyDeleteFrequency sampling realization is computationally more efficient than direct form realization
ReplyDelete