A modified svd multi-frequency quantitative inversion method for weak scatterers

In this paper, a modified singular value decomposition (SVD) method for quantitative inversion of the weak scatterers with multi-frequency data is presented. The proposed method can utilize the low frequency information content effectively and obtain the permittivity and the conductivity simultaneously. Meanwhile, multi-frequency configurations based on the equal wavelength interval and the equal frequency interval are discussed for more accurate inversion. The modified SVD is proposed by firstly reconstructing the scattered field and the operator, then decoupling the permittivity and conductivity, and finally mapping the scattered field to the contrast function. Inversion results with synthetic data demonstrate that the proposed method outperforms the conventional SVD approach in terms of quantitative inversion accuracy for weak scatterers of the multi-frequency data.