A neural-network approach to the electromagnetic imaging of elliptic conducting cylinders

Abstract

A neural-network approach is presented for the electromagnetic imaging of conducting cylinders of elliptical cross section. Simulation results are presented which demonstrate that a properly trained neural network is capable of accurately identifying the shape and location of unknown elliptic cylinders in inaccessible domains. © 2001 John Wiley & Sons, Inc. Microwave Opt Technol Lett 28: 303–306, 2001.