Conditions for generation of stable and accurate hybrid TD-FD MoM solutions

Abstract

Broadband characterization of any electromagnetic (EM) data (e.g., surface currents, radiation pattern, and network parameters) can be carried out using partial information in the time domain (TD) and the frequency domain (FD). In this hybrid TD-FD approach, one generates the early time response using a TD code at a spatial location and uses a FD code to generate the low-frequency response at the same place. Then, the partial complementary information in both the TD and FD is fit by a set of orthogonal functions and its Fourier transform having the same expansion coefficients. Three different types of functions, namely, Hermite, Bessel-Chebyshev, and Laguerre, have been used for extrapolation. Once the expansion coefficients for these functions are known, the response can be extrapolated either for late times or high frequencies using the initial partial information. The objective of this paper is to explore the conditions under which this hybrid approach yields a stable and accurate solution. We investigate bounds for both the number of orthogonal functions needed to carry out the extrapolation and the scale factors needed to accurately fit the data in time and in frequency. Numerical examples have been presented to illustrate the efficacy of these bounds. It is important to point out that, in this hybrid approach of extrapolation, we are not creating new information but processing the available information in an intelligent fashion.Index Terms-Bessel-Chebyshev, frequency domain (FD), Hermite, hybrid method, Laguerre, lower bounds, orthogonal polynomials, scaling factor, time domain (TD), upper bounds.

Implementation of Associated Hermite FDTD Method in Handling INBCs for Shielding Analysis

“…One can also calculate from the frequency-domain data by using isomorphism property of AH function [21]. For a proper approximation with AH expansion coefficients, the parameters selection can be selected according to [20,24]:…”

A New Unconditionally Stable Scheme for FDTD Method Using Associated Hermite Orthogonal Functions

“…The time support and frequency support of the AH functions can be approximated by the following empirical formula [24] (1)…”

“…If the magnetic field vector variables in the whole computational domain are calculated, the electronic field vector variables and can be obtained from ( 20)- (21). Rewriting (23) as a nested matrix equation, we have (24) where is an unknown vector assembled with vector variable elements that is consist of magnetic field coefficients of all orders for all points.…”

Image watermarking based on the space/spatial-frequency analysis and Hermite functions expansion

“…According to the masking effects of the human visual system (HVS), the busy or high contrast image areas can mask any distortion better than smooth areas, and thus, can support a higher watermark signal without visibility problem. [23][24][25][26] Hence, when using a few low-order functions for the reconstruction, the MSE for busy regions will be pretty high, since these regions are characterized by high-frequency content. 21,22 The similar design 14 uses the energy of local space/spatialfrequency content to determine the occupancy and busyness of the region.…”