Fast numerical integration of large planar apertures using a sparse expansion of the far-field pattern

The application of compressive sensing to computational electromagnetics is demonstrated by considering the radiation from large polygon-shaped apertures. For a well-behaved aperture distribution, it is known that the radiation may be is found in terms of virtual sources located at the vertices. This sparse representation enables the use of sub-Nyquist sampling to obtain the coefficients of the expansion. Two orders of magnitude reduction in the number of sampled pattern points is demonstrated.