Butterworth Filter in C#

public static double[] Butterworth(double[] indata, double deltaTimeinsec, double CutOff) {
    if (indata == null) return null;
    if (CutOff == 0) return indata;

    double Samplingrate = 1 / deltaTimeinsec;
    long dF2 = indata.Length - 1;        // The data range is set with dF2
    double[] Dat2 = new double[dF2 + 4]; // Array with 4 extra points front and back
    double[] data = indata; // Ptr., changes passed data

    // Copy indata to Dat2
    for (long r = 0; r < dF2; r++) {
        Dat2[2 + r] = indata[r];
    }
    Dat2[1] = Dat2[0] = indata[0];
    Dat2[dF2 + 3] = Dat2[dF2 + 2] = indata[dF2];

    const double pi = 3.14159265358979;
    double wc = Math.Tan(CutOff * pi / Samplingrate);
    double k1 = 1.414213562 * wc; // Sqrt(2) * wc
    double k2 = wc * wc;
    double a = k2 / (1 + k1 + k2);
    double b = 2 * a;
    double c = a;
    double k3 = b / k2;
    double d = -2 * a + k3;
    double e = 1 - (2 * a) - k3;

    // RECURSIVE TRIGGERS - ENABLE filter is performed (first, last points constant)
    double[] DatYt = new double[dF2 + 4];
    DatYt[1] = DatYt[0] = indata[0];
    for (long s = 2; s < dF2 + 2; s++) {
        DatYt[s] = a * Dat2[s] + b * Dat2[s - 1] + c * Dat2[s - 2]
                   + d * DatYt[s - 1] + e * DatYt[s - 2];
    }
    DatYt[dF2 + 3] = DatYt[dF2 + 2] = DatYt[dF2 + 1];

    // FORWARD filter
    double[] DatZt = new double[dF2 + 2];
    DatZt[dF2] = DatYt[dF2 + 2];
    DatZt[dF2 + 1] = DatYt[dF2 + 3];
    for (long t = -dF2 + 1; t <= 0; t++) {
        DatZt[-t] = a * DatYt[-t + 2] + b * DatYt[-t + 3] + c * DatYt[-t + 4]
                    + d * DatZt[-t + 1] + e * DatZt[-t + 2];
    }

    // Calculated points copied for return
    for (long p = 0; p < dF2; p++) {
        data[p] = DatZt[p];
    }

    return data;
}