Frequency-Dependent Modeling of Ultra-WideBand Pulses in Human Tissue for Biomedical Applications

Due to the recent advances in ultra wide-band (UWB) radar technologies, there has been widespread interest in the possible medical applications of UWB microwave radar. To facilitate the development of signal processing algorithms, numerical methods may be used to generate representative backscattered signals. Two techniques for modelling the propagation of electromagnetic waves (UWB pulses) in human tissue, are presented: the planar technique and finite difference time domain (FDTD) technique. A four layer biological model is considered with three layers of normal tissue, and one layer of cancerous soft tissue (sarcoma). The two modelling techniques are used to predict the response of the model to the UWB input signal, with particular emphasis on the response of the sarcoma layer. It will be shown that both the Planar technique and the FDTD technique identify the presence of the soft tissue sarcoma quite easily. However the FDTD technique predicts more subtle phenomena such as multiple reflections, albeit at a higher computational cost