Abstract

The mathematical treatment involving Fresnel diffraction in compound linear optical systems presents difficulties due to the high degree of oscillations of the Fresnel integral kernel. Also, it is necessary to impose small pupil dimensions on the elements forming the compound system. A revised formulation for the general system having an arbitrary number of elements and generalized pupil functions allows one to represent the kernel in terms of convolution operations. This leads to a compact representation using Lie algebra operators.

© 1989 Optical Society of America