【Stockham FFT の種類】
ここでは、私が Web を検索して見つけた Stockham のアルゴリズムの種類について整理しておきます。まず、 このサイトで説明されている Cooley-Tukey のアルゴリズムからの単純な変形で導ける Stockham のアルゴリズムをタイプ1。 私がなぜこれでフーリエ変換できるのか説明できない Stockham のアルゴリズムをタイプ2とします。
タイプ1、 タイプ2それぞれに周波数間引き(DIF)と時間間引き(DIT)のアルゴリズムがあります。 すなわち、Stockham のアルゴリズムは最低でも4種類あることになります。
タイプ1-DIF Stockham アルゴリズム
#include <complex> #include <cmath> typedef std::complex<double> complex_t; void fft0(int n, int s, bool eo, complex_t* x, complex_t* y) { const int m = n/2; const double theta0 = 2*M_PI/n; if (n == 1) { if (eo) for (int q = 0; q < s; q++) y[q] = x[q]; } else { for (int p = 0; p < m; p++) { const complex_t wp = complex_t(cos(p*theta0), -sin(p*theta0)); for (int q = 0; q < s; q++) { const complex_t a = x[q + s*(p + 0)]; const complex_t b = x[q + s*(p + m)]; y[q + s*(2*p + 0)] = a + b; y[q + s*(2*p + 1)] = (a - b) * wp; } } fft0(n/2, 2*s, !eo, y, x); } } void fft(int N, complex_t* x) { complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] /= N; } void ifft(int N, complex_t* x) { for (int p = 0; p < N; p++) x[p] = conj(x[p]); complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] = conj(x[k]); }
タイプ1-DIT Stockham アルゴリズム
#include <complex> #include <cmath> typedef std::complex<double> complex_t; void fft0(int n, int s, bool eo, complex_t* x, complex_t* y) { const int m = n/2; const double theta0 = 2*M_PI/n; if (n == 1) { if (eo) for (int q = 0; q < s; q++) x[q] = y[q]; } else { fft0(n/2, 2*s, !eo, y, x); for (int p = 0; p < m; p++) { const complex_t wp = complex_t(cos(p*theta0), -sin(p*theta0)); for (int q = 0; q < s; q++) { const complex_t a = y[q + s*(2*p + 0)]; const complex_t b = y[q + s*(2*p + 1)] * wp; x[q + s*(p + 0)] = a + b; x[q + s*(p + m)] = a - b; } } } } void fft(int N, complex_t* x) { complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] /= N; } void ifft(int N, complex_t* x) { for (int p = 0; p < N; p++) x[p] = conj(x[p]); complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] = conj(x[k]); }
タイプ2-DIF Stockham アルゴリズム
#include <complex> #include <cmath> typedef std::complex<double> complex_t; void fft0(int n, int s, bool eo, complex_t* x, complex_t* y) { const int m = n/2; const double theta0 = M_PI/s; if (n == 1) { if (eo) for (int q = 0; q < s; q++) x[q] = y[q]; } else { fft0(n/2, 2*s, !eo, y, x); for (int p = 0; p < m; p++) { for (int q = 0; q < s; q++) { const complex_t wq = complex_t(cos(q*theta0), -sin(q*theta0)); const complex_t a = y[q + s*(2*p + 0)]; const complex_t b = y[q + s*(2*p + 1)]; x[q + s*(p + 0)] = a + b; x[q + s*(p + m)] = (a - b)*wq; } } } } void fft(int N, complex_t* x) { complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] /= N; } void ifft(int N, complex_t* x) { for (int p = 0; p < N; p++) x[p] = conj(x[p]); complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] = conj(x[k]); }
タイプ2-DIT Stockham アルゴリズム
#include <complex> #include <cmath> typedef std::complex<double> complex_t; void fft0(int n, int s, bool eo, complex_t* x, complex_t* y) { const int m = n/2; const double theta0 = M_PI/s; if (n == 1) { if (eo) for (int q = 0; q < s; q++) y[q] = x[q]; } else { for (int p = 0; p < m; p++) { for (int q = 0; q < s; q++) { const complex_t wq = complex_t(cos(q*theta0), -sin(q*theta0)); const complex_t a = x[q + s*(p + 0)]; const complex_t b = x[q + s*(p + m)] * wq; y[q + s*(2*p + 0)] = a + b; y[q + s*(2*p + 1)] = a - b; } } fft0(n/2, 2*s, !eo, y, x); } } void fft(int N, complex_t* x) { complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] /= N; } void ifft(int N, complex_t* x) { for (int p = 0; p < N; p++) x[p] = conj(x[p]); complex_t* y = new complex_t[N]; fft0(N, 1, 0, x, y); delete[] y; for (int k = 0; k < N; k++) x[k] = conj(x[k]); }