Application of signal-subspace and optimization methods in reconstructing extended scatterers

Area of Science:

• Electromagnetics and Wave Propagation
• Inverse Problems
• Computational Physics

Background:

• Accurate reconstruction of material properties (permittivities) is crucial for characterizing extended scatterers.
• Traditional methods may struggle with noise, convergence speed, and data requirements.
• Two-dimensional inverse scattering problems present unique challenges in computational electromagnetics.

Purpose of the Study:

• To propose a novel signal-subspace approach for reconstructing the permittivities of extended scatterers in 2D.
• To enhance the accuracy and efficiency of inverse scattering algorithms.
• To address limitations of existing methods regarding noise and data acquisition.

Main Methods:

• A hybrid approach combining a signal-subspace method with a nonlinear least-squares problem.
• Utilizing a portion of scatterer information via the signal-subspace technique.
• Solving the remaining reconstruction problem iteratively using least-squares optimization.

Main Results:

• The proposed method successfully reconstructs scatterer permittivities with high resolution.
• Demonstrated robustness against noise in scattering data.
• Achieved fast convergence and required less scattering data compared to conventional techniques.
• Successfully handled scatterers with complex or special shapes.

Conclusions:

• The signal-subspace approach provides an effective and efficient solution for 2D inverse scattering problems.
• This method offers significant advantages in terms of accuracy, speed, and robustness.
• The technique is well-suited for practical applications requiring detailed material property reconstruction.