Simulation of Magnetic Induction Tomography Sensor with 8-Coils Solenoid and Planar

Abstract

Magnetic Induction Tomography (MIT) is a tomography method that utilizes magnetic field induced by coils excited with an electric current. A conductive object placed within the Region of Interest (ROI) will cause a perturbation in the magnetic field and cause a voltage difference measured in the sensor. These voltage data will be processed by implementing the Linear Back-Projection (LBP) algorithm to reconstruct the conductivity distribution image with a resolution of $32\text{x}32$ pixels. Simulation will be used as a method to obtain the sensor's characteristic in the form of a sensitivity matrix. Voltage measurements when a conductive object is placed within the ROI can also be acquired through simulation done with the COMSOL® Multiphysics® 4.4 software. The obtained data will be processed using the MATLAB R2013a software. Based on the results of the MIT sensor simulation, it is retrieved that although the LBP algorithm is a fast and simple method to implement in the MIT system, it can sense a conductive object placed in the ROI though it is still a very rough estimate to map conductivity distribution.