Open eblur opened 8 years ago
Instead of calculating the bessel function terms each time, use caching to store values so there are fewer repeated calculations.
Input terms: a (grain size), E (photon energy), n (index) -- see Van de Hulst or Bohren & Huffman
Compute S1 and S2 matrices of dimension a, E, n. Add along n-axis any time a new term is needed.
See branch speed_up_mie
eblur
commented
8 years ago eblur
commented
8 years ago
Instead of calculating the bessel function terms each time, use caching to store values so there are fewer repeated calculations.
Input terms: a (grain size), E (photon energy), n (index) -- see Van de Hulst or Bohren & Huffman
Compute S1 and S2 matrices of dimension a, E, n. Add along n-axis any time a new term is needed.