A Hybrid Tomographic Reconstruction for Dielectric-varying Scatterers Based on Kalman Filtering and Distorted Born Iterative Method

This paper presents a novel hybrid microwave tomographic method based on Kalman filter and Distorted Born Iterative Method (DBIM) for scatterers that behave time-varying permittivity. Unlike conventional microwave tomographic techniques where the scatterer is regarded as an object with a constant permittivity, the proposed method can be applied to the scenarios where the scatterer has its permittivity changing with time (e.g. to monitor the recovering of bone fractures the permittivity of the diseased area is decreasing). In this hybrid method, the solution of DBIM is implemented as measurements of Kalman filter, whereas the permittivity and its changing speed are taken as the states of Kalman filter. Consequently, the optimal estimation can be used to estimate the optimal result. Finally, this method is evaluated on a numerical scatterer model. Compared to the existing imaging reconstruction methods, it shows that the proposed hybrid method can locate the permittivity-varying scatterer accurately.