Experimental investigation of an optimization approach to microwave tomography

Microwave imaging is an interesting and growing research field with a number of medical applications. This paper is based on the first series of experimental results using an iterative gradient algorithm based on the finite difference time domain (FDTD) method and synthetic pulses. Using our method, the permittivity and conductivity of an object are reconstructed layer by layer by minimizing a functional consisting of the difference between the measured and calculated electric field surrounding the object. This is done by surrounding the object with a number of antennas which are all in turn transmitting and receiving. The dielectric profiles used in the calculations are then iteratively updated until the functional is minimized. Results are presented demonstrating the ability to detect metallic and dielectric material in air and water.