论文信息 - Application of neural networks to inverse problems in electromagnetics

Application of neural networks to inverse problems in electromagnetics

The paper presents a neural network approach for solving inverse problems in electromagnetics. In general, the measurements of the scattered electromagnetic fields can be related to the properties of the scatterer through integral equations. The energy minimization property of Hopfield neural networks is exploited to solve these integral equations. Results of implementing the method on an application problem involving reconstruction of material properties of multilayered media are presented. >

Satish S. Udpa | Lalita Udpa | I. Elshafiey

[1] Weng Cho Chew,et al. Response of a current loop antenna in an invaded borehole , 1984 .

[2] Weng Cho Chew,et al. Simultaneous reconstruction of permittivity and conductivity profiles in a radially inhomogeneous slab , 1986 .

[3] Curtis R. Vogel,et al. Numerical solution of a non-linear ill-posed problem arising in inverse scattering , 1985 .

[4] K. Mei,et al. Inverse scattering technique applied to remote sensing of layered media , 1981 .

[5] J J Hopfield,et al. Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[6] John J. Hopfield,et al. Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[7] Fathi M. A. Salam,et al. On the analysis of dynamic feedback neural nets , 1991 .