Title :
Frequency selective surface design using neural networks inversion based on parametrized representations
Author :
Davis, D.T. ; Chan, C.H. ; Hwang, J.N.
Author_Institution :
Dept. of Electr. Eng., Washington Univ., Seattle, WA, USA
Abstract :
A parametric model of a frequency selective surface (FSS) is presented. By using a parametric representation of the FSS, one can simplify the process of designing an FSS for a given response by embedding constraints into the input data representation, thus avoiding the need for the constraint satisfaction mechanism. A parametric representation of an FSS made up of a dipole array is considered as an example.<>
Keywords :
antenna arrays; antenna theory; dipole antennas; neural nets; FSS; dipole array; frequency selective surface; input data representation; neural networks inversion; parametric model; parametric representation; Algorithm design and analysis; Apertures; Error correction; Frequency response; Frequency selective surfaces; Iterative algorithms; Multilayer perceptrons; Neural networks; Neurons; Surface treatment;
Conference_Titel :
Antennas and Propagation Society International Symposium, 1991. AP-S. Digest
Conference_Location :
London, Ontario, Canada
Print_ISBN :
0-7803-0144-7
DOI :
10.1109/APS.1991.174807
