SCATTERING FROM GROOVE PATTERNS IN A PERFECTLY CONDUCTING SURFACE

Electromagnetic scattering is investigated for assemblages of parallel open cavities recessed in a perfectly conducting ground plane. Caviti es of a variety of shapes are treated, with cross-sectional dimensions of the order of one or two electromagnetic wavelengths. Under the ass umption that the cavities form grooves of effectively infinite length, a two-dimensional analysis treats transverse incidence under both E-a nd H-polarized illumination (E and H fields parallel to groove axis, r espectively). For the most part, any coupling between cavity responses on the surface produces negligible effects on far-field diffraction p atterns, even when cavities are extremely close together and when indu ced currents flow between adjacent cavities. Thus one may usually cons truct diffraction patterns for assemblages of grooves by simply superp osing responses calculated for each cavity in isolation. Despite possi bly substantial differences among the individual scattering patterns f rom contributing cavities, regularly spaced arrangements of two or mor e cavities produced grating-type diffraction patterns. This allows inf erence of the distance between grooves, based on separation between th e pattern's peaks and troughs. Combinations of dissimilar cavities may produce diffraction patterns with peaks that are shifted away from lo cations expected on the basis of the grating equation, but with a char acteristic spacing between the peaks approximately preserved. Random p erturbation of groove locations relative to uniform spacing produces a decay in ensemble-average diffraction pattern as scattering angles di verge from the specular direction. A simple theory quantifies the expo nential dependency of this grating pattern suppression and shows its e ffect on the angular range available for identification of scattering characteristics. Monte Carlo-type backscatter simulations including al l intercavity coupling demonstrate the success of the theory and there by explain the type of pattern suppression seen in measurements. (C) 1 997 Optical Society of America.