Examination, Clarification, and Simplification of Modal Decoupling Method for Multiconductor Transmission Lines

Guang Tsai Lei, Barry K. Gilbert, Guang Wen (George) Pan

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


In the application of the modal decoupling method, questions arise as to why the nonnormal matrices LC and CL are diagonalizable. Is the definition of the characteristic impedance matrix Zc unique? Is it possible to normalize current and voltage eigenvectors simultaneously, yet assure the correct construction of the Zc matrix? Under what conditions do Mit Mv = I and Zc = Mv Mi-1? In this paper, these questions are thoroughly addressed. We will prove the diagonalizability of matrices LC and CL for lossless transmission lines (though the diagonalizability of their complex analogues, ZY and YZ matrices, is not guaranteed for lossy lines), and will demonstrate the properties of their eigenvalues. We have developed an algorithm to decouple one type of matrix differential equation, and to construct the characteristic impedance matrix Zc explicitly and efficiently. Based on this work, the congruence and similarity transformations, which have caused considerable confusion and not a few errors in the decoupling and solution of the matrix telegrapher's equations, will be analyzed and summarized. In addition, we will also demonstrate that under certain conditions, the diagonalization of two or more matrices by means of the congruence or similarity transformations may lead to coordinate system “mismatch” and introduce erroneous results.

Original languageEnglish (US)
Pages (from-to)2090-2100
Number of pages11
JournalIEEE Transactions on Microwave Theory and Techniques
Issue number9
StatePublished - Sep 1995
Externally publishedYes

ASJC Scopus subject areas

  • Radiation
  • Condensed Matter Physics
  • Electrical and Electronic Engineering


Dive into the research topics of 'Examination, Clarification, and Simplification of Modal Decoupling Method for Multiconductor Transmission Lines'. Together they form a unique fingerprint.

Cite this