| Abstract: | Discrete Fourier transform analysis provides an infinite number of weight coefficients for filters like upward and downward continuation. For practical applicability, the lengths of such filters have been reduced to a manageable number by various shortening operators, viz. those by Peters, Martin, Mufti, v. Hann, Hamming, and the truncation operator. A comparative study for choosing an operator which approximates the theoretical filter response best has indicated that Martin's shortening operator and the truncation operator are best, respectively, for normalized and non-normalized sets of weight coefficients. |
