Shape Reconstruction of Dielectric and Conducting Objects using Linear Sampling Method and Limitations

This paper aims to investigate the behaviour of linear inverse method, linear sampling method (LSM), to estimate geometric properties of dielectric/ conducting objects. LSM is a simple, reliable linear inverse algorithm and uses multiview multistatic scattered field data measured around target objects. Despite its simplicity and computational effectiveness, LSM fails to detect targets support at certain frequencies (eigenvalues). The present work focusses on the occurrence of eigenvalues with respect to the object geometric properties. All these analyses are done by considering the numerical examples of synthetic data as well as experimental data provided by Institute of Fresnel.