General Theory of Coupled Characteristic Mode: An Eigen Subspace Approach
Rakesh Sinha, Sandip Ghosal · 2025
Read General Theory of Coupled Characteristic Mode: An Eigen Subspace Approach by Rakesh Sinha, Sandip Ghosal (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-08251.txt.
Introduction
Rakesh Sinha and Sandip Ghosal present a 2025 theory of coupled characteristic modes for antennas and scatterers. Full eigenanalysis of the impedance matrix is costly. In practice a few modes describe a device well. Their eigen-subspace approach keeps only that useful slice.
They assume the coupled mode space matches the uncoupled mode plane. Coupled modes then read as mixes of isolated modes. A modal coupling matrix encodes the mixing. That matrix explains behavior for odd shapes without full solves. The shortcut could speed antenna design loops a great deal. It also cuts computing cost for large arrays.
Catalog record
| CREATOR | Rakesh Sinha, Sandip Ghosal |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-08251
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
General Theory of Coupled Characteristic Mode: An Eigen Subspace Approach Rakesh Sinha, Sandip Ghosal Abstract In this work, the problem of characteristic mode analysis using eigendecomposition of the method of moments impedance matrix has been simplified using the eigen-subspace approach. The idea behind the eigen-subspace arises from the physical properties of antenna or scatterers, where only a few eigenmodes are enough to characterize the antenna or scatterer. Therefore, entire space eigenanalysis is a waste of computational resources, and eigen-subspace analysis with few modes is good enough to characterize antennas and scatterers. It has been assumed that there is an eigen-subspace (or hyperplane) of coupled characteristic mode, which coincides with the eigen- hyperplane of uncoupled characteristic mode. We can say the coupled characteristic modes are linear combinations of isolated modes based on this assumption. The linear combination is mapped via modal coupling matrix . Using the modal coupling matrix , we can explain the behavior of arbitrarily shaped antennas and scatterers. A computationally efficient method is developed to compute coupled characteristic modes of two mutually coupled scatterers or antennas using the eigen- subspace. The method is summarized as a theorem of two-body coupled characteristic mode. The theorem of two-body coupled characteristic mode has been extended to the N-body coupled characteristic mode. Two algorithms have been developed for the two-body multimode coupled characteristic mode and N-body multimode coupled characteristic mode. Two numerical examples are provided to validate the proposed concepts. Nomenclature {[\textbf{Z}]_{N\times N}} Method of moments Z-matrix with N basis function. {[\textbf{R}]_{N\times N}} and {[\textbf{X}]_{N\times N}} Re. and Im. part of MoM self impedance matrix [\textbf{Z}]=[\textbf{R}]+j[\textbf{X}] . {[\text{V}]_{N\times 1}} and {[\text{I}]_{N\times 1}} Excitation vector and surface current coefficient vector in MoM formulation [\textbf{Z}][\text{I}]=[\text{V}] . {[\textbf{I}]_{N\times N}} =[I_{1},I_{2},..,I_{N}] : Set of N eigenvectors covers entire eigenspace of characteristic mode equation ( I_{k}:(\textbf{X}-\lambda_{k}\textbf{R})I_{k}=0 ) with [I_{k}] is the k^{th} characteristic mode. {[\bm{\lambda}]_{N\times N}} =diag(\lambda_{1},\lambda_{2},..,\lambda_{N}) : All the eigenvalues of characteristic mode equation with \lambda_{k} is the k^{th} eigenvalue. {[\underaccent{\bar}{\textbf{I}}]_{N\times k}} =[I_{1},I_{2},..,I_{k}] : Set of first k eigenvectors covers an eigen-subspace or hyperplane span \{I_{1},I_{2},..,I_{k}\} of characteristic mode equation. {[\underaccent{\bar}{\bm{\lambda}}]_{k\times k}} =diag(\lambda_{1},\lambda_{2},..,\lambda_{k}) : First k eigenvalues of characteristic mode equation. {[\underaccent{\bar}{\textbf{P}}_{Z}]_{k\times k}} =[\underaccent{\bar}{\textbf{I}}]^{T}[\textbf{Z}][\underaccent{\bar}{\textbf{I}}] : Complex modal power matrix of first k modes. {[\underaccent{\bar}{\textbf{P}}_{R}]_{k\times k}} =diag(P_{R1},P_{R2},...,P_{Rk}) : Modal radiated power matrix of first k modes with P_{Rk}=[I_{k}]^{T}[\textbf{R}][I_{k}] . {[\underaccent{\bar}{\textbf{P}}_{X}]_{k\times k}} =diag(P_{X1},P_{X2},...,P_{Xk}) : Modal reactive power matrix of first k modes with P_{Xk}=[I_{k}]^{T}[\textbf{X}][I_{k}] . {[\underaccent{\bar}{\text{E}}]} =[e_{1},e_{2},..e_{k}]^{T} : Modal excitation of first k modes with e_{k}=[I_{k}]^{T}[\text{V}]. {[\alpha]} =[\underaccent{\bar}{\textbf{P}}_{Z}]^{-1}[\underaccent{\bar}{\text{E}}] : Modal contribution of first k modes in total surface current [\text{I}]\approx[\underaccent{\bar}{\textbf{I}}][\alpha] . {[\textbf{Z}_{AA}]} and {[\textbf{Z}_{BB}]} MoM self impedance matrix of A and B. {[\textbf{Z}_{AB}]} =[\textbf{R}_{AB}]+j[\textbf{X}_{AB}] : MoM transfer (coupling) impedance matrix between A and B. {[\underaccent{\bar}{\textbf{P}}_{Z,AA}]} and {[\underaccent{\bar}{\textbf{P}}_{Z,BB}]} The complex (radiated and reactive) modal power matrix (diagonal) first k modes of A and l modes of B. {[\underaccent{\bar}{\textbf{P}}_{Z,AB}]} =[\underaccent{\bar}{\textbf{P}}_{R,AB}]+j[\underaccent{\bar}{\textbf{P}}_{X,AB}] : The complex (radiated and reactive) modal transfer (coupling) power matrix between first k modes of A and l modes of B. {[\textbf{Z}^{uc}]} =[\textbf{Z}_{AA},\textbf{0};\textbf{0},\textbf{Z}_{BB}] : MoM impedance matrix of the uncoupled system. {[\underaccent{\bar}{\bm{\lambda}}^{uc}]} =[\underaccent{\bar}{\bm{\lambda}}_{A},\bm{0};\bm{0},\underaccent{\bar}{\bm{\lambda}}_{B}] : The set of eigenvalues of the uncoupled system [\textbf{Z}^{uc}] consist of k modes of [\textbf{Z}_{AA}] and l modes of [\textbf{Z}_{BB}] . {[\underaccent{\bar}{\textbf{I}}^{uc}]} =[\underaccent{\bar}{\textbf{I}}_{A},\bm{0};\bm{0},\underaccent{\bar}{\textbf{I}}_{B}] : The set of eigenvectors of the uncoupled system [\textbf{Z}^{uc}] consist of k modes of [\textbf{Z}_{AA}] and l modes of [\textbf{Z}_{BB}] . {[\textbf{Z}^{c}]} =[\textbf{Z}_{AA},\textbf{Z}_{AB};\textbf{Z}_{BA},\textbf{Z}_{BB}] : MoM impedance matrix of the coupled system. {[\underaccent{\bar}{\bm{\lambda}}^{c}]} and {[\underaccent{\bar}{\textbf{I}}^{c}]} The sets of first k+l eigenvalues and eigenvectors pairs of the coupled system [\textbf{Z}^{c}] consist of linear combinations k modes of [\textbf{Z}_{AA}] and l modes of [\textbf{Z}_{BB}] . {[\underaccent{\bar}{\bm{\lambda}}^{c}_{A(B)}]} and {[\underaccent{\bar}{\textbf{I}}^{c}_{A(B)}]} The sets of first k or l eigenvalues and eigenvectors pairs of the coupled system [\textbf{Z}^{c}] have major contributions from k modes of [\textbf{Z}_{AA}] or l modes of [\textbf{Z}_{BB}] . {[\underaccent{\bar}{\textbf{M}}]_{(k+l)\times(k+l)}} Modal coupling matrix represents modal coupling between k modes of A and l modes B with [\underaccent{\bar}{\textbf{I}}^{c}]=[\underaccent{\bar}{\textbf{I}}^{uc}][\underaccent{\bar}{\textbf{M}}] . I Introduction The electromagnetic radiation and scattering phenomena of arbitrary objects are analyzed using the equivalent current vector. With the gradual demand of tightly coupled electromagnetic components within an electrically smaller real estate, it becomes challenging to determine asymptotic or analytic solution [ 1 ] –- [ 4 ] of such coupled system. This inspires to explore numerical techniques like method of moments (MoM) [ 5 ] and [ 6 ] . Although computational advancement makes it possible to solve complex coupled structures numerically, it only provides a set of numbers with little insight into the underlying physics. It motivates to the development of various modal approaches like [ 7 ] and [ 8 ] . For common geometries like spheres or cylinders, there exist spherical or cylindrical wave functions to analyze electromagnetic behavior. This provides the initial intuition of characteristic wave functions for arbitrarily shaped objects in [ 7 ] and [ 8 ] . Similar to the closed-form solution of the resonating modes of commonly shaped wave-guide like structures, a mathematical representation of the resonant scattering phenomena was initially proposed in [ 7 ] and [ 8 ] using the eigenvectors of the object’s far-field scattering matrix. Later, the electric field integral equation (EFIE) was followed to determine the characteristic current modes through generalized eigen decomposition (GED) of the MoM impedance matrix in [ 9 ] – [ 11 ] . Although the electric/magnetic current-based formulation of characteristic mode analysis (CMA) was initially introduced in the seventies, the lack of suitable computing system restricted its popularity till the nineties. With the development of computing facilities and matrix-based platforms like MATLAB [ 12 ] , resurgent interests in CMA were noted for the antenna design and EMI/EMC applications [ 13 ] – [ 22 ] . It is to be noted that the computational complexities of the matrix inversion operation of conventional MoM and generalized eigendecomposition operation are the same and each follows the cubic order of the MoM impedance matrix size [ 23 ] , unless any iterative algorithm is used. As the number of elements increases in a coupled array, the size of the MoM impedance matrix also increases proportionately. As a consequence, the conventional CMA technique of [ 9 ] and [ 10 ] urges for a high computational load. This leads to the search for various alternate techniques to determine the characteristic modes of the coupled array. Initial numerical experimentation on this topic was reported in [ 13 ] and [ 15 ] where the modal behavior was studied in the resonating region for the linear arrays of electric and magnetic dipoles. Later, the matrix partitioning method was used to analyze two-element and three-element arrays in [ 24 ] . For a two-element array with excitation in only one element, a different set of eigenmodes was defined as the substructure characteristic modes (SCM) in [ 25 ] . Similarly, a lower-order matrix operation was proposed for the finite array of identical elements in [ 26 ] . Another formulation was reported recently for the CMA of finite periodic arrays in [ 27 ] . A rigorous matrix order reduction method was reported in [ 28 ] for finite symmetric arrays which provides a Toeplitz nature in the MoM impedance matrix. The behavior of mode coupling between PEC and lossy dielectric was studied