Abstract
Conference Title: 2018 IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting Conference Start Date: 2018, July 8 Conference End Date: 2018, July 13 Conference Location: Boston, MA, USA In this paper, the Poggio-Miller-Chang-Harrington-Wu-Tsai Equations associated with the Mixed Potential Integral Equations (PMCHWT-MPIE) are used to analyze objects comprising of lossy composite materials embedded in layered media, and the Discrete Complex Image Method (DCIM) is used to generate the Dyadic Green's Functions. Compared to the MOM analysis based on the Impedance Boundary Condition (IBC), the PMCHWT-based analysis is more robust when dealing with highly lossy objects embedded in layered media. We choose the Characteristic Basis Functions Method (CBFM) for this problem because it is iteration-free and, hence, is suitable for dealing with multiple RHS efficiently. Yet another reason for this choice is that the GPU can be used to accelerate the filling process of the submatrices of the impedance matrix when generating the Characteristic Basis Functions (CBFs).