On-line inverse scattering of conducting cylinders using radial basis-function neural networks

A novel on-line inverse-scattering method for the geometric characterizatio n of conducting cylindrical scatterers from scattered-field measurements is presented. The method is based on the application of radial basis-function neural networks that are constructed by use of the orthogonal least square s algorithm. Thus, trial-and-error approaches during the construction can b e avoided. The efficiency and robustness of the method in inverting noisy m easurements are investigated. (C) 2001 John Wiley & Sons, Inc.