besi0 besi0e besi1 besi1e besin besind besinu besinud besk0 besk0e besk1 besk1e beskn besknd besknu besknud cbesi cbesk sbesin sbesinu sbeskn sbesknu

cbesk() void cbesk ( doublecomplex  z, double  nu, int  kode, int  n, doublecomplex  y[], int *  info  ) Sequence of the modified Bessel functions of the second kind Kν(z) (fractional order) (complex argument) Purpose cbesk computes the sequence of the modified Bessel functions of the second kind Kν(z) for complex nonzero argument z and real nonnegative orders ν, ν+1, ... A scaling option is available to help avoid overflow. Parameters [in] z Argument z. (z != 0) [in] nu Order of first member of the sequence. (nu >= 0) [in] kode A parameter to indicate the scaling option. = 1: No scaling. = 2: Returns exponentially scaled values. [in] n Number of members in the sequence. (n members of orders nu to nu+n-1 to be computed) (n >= 1) [out] y[] Array y[ly] (ly >= n) Sequence of Bessel functions of the second kind. kode = 1: Kν(z) (ν = nu to nu+n-1) kode = 2: exp(z)*Kν(z) (ν = nu to nu+n-1) [out] info = 0: Successful exit. = -1: The argument z had an illegal value. (z = 0) = -2: The argument nu had an illegal value. (nu < 0) = -3: The argument kode had an illegal value. (kode != 1 and kode != 2) = -4: The argument n had an illegal value. (n < 1) = nz (0 < nz <= n): The last nz components of y[] set to zero due to underflow. = n+1: Precision warning. (Half precision or less because abs(z) or nu+n-1 is large) = n+2: Precision error. (No precision because abs(z) or nu+n-1 is too large) = n+3: Overflow abs(z) too small and/or nu+n-1 too large. = n+4: Algorithmic error. (Termination condition not met) Reference SLATEC