Connect with us

Evaluation of the square root of a complex matrix

Discussion in 'Electronic Basics' started by Julian Grodzicky, Sep 28, 2013.

Scroll to continue with content
  1. Summary: How do I evaluate the square root of a complex matrix?

    I am studying from Pieter LD Abrie's "Design of RF and Microwave Amplifiersand Oscillators", 2ed, Artech House, 2009.

    In the first chapter he presents derivations of S-parameters for N-port networks (sec. 1.5), he states an expression on page 13, (sqrt(2))^(-1)*[Z_0 +Z_0*]^(1/2), where the matrix of impedances is Z_0 = [R_0i +
    jX­­_0i], and Z_0* is the complex conjugate matrix of Z_0; j=sqrt(-1); R_0i is the ith resistance of port i and X_0i is the ith reactance of port i.

    I'm trying to derive this expression, but all the texts on Linear Algebra that I have in my personal library only deal with positive integer powers ofreal matrices. (I have in my arsenal "Elementary Linear Algebra: Applications Version", H.Anton & C.Rorres, 6ed, Wiley: "Schaum's Outline of Linear Algebra; "Mathematical Methods for Physics and Engineering", 2ed, K.F.Riley, M.P.Hobson & S.J.Bence, Cambridge UP).

    What topics and texts do I need to study to be able to evaluate the square root of a complex matrix?

    Cheers,
    Julian

    PS Anyone know of a forum where they provide/allow mathematical notation?
     
  2. o pere o

    o pere o Guest

    If you just want to numerically evaluate a result, you may simply use
    octave (or Matlab)

    octave:1> sqrt([1 j;2 j])
    ans =

    1.00000 + 0.00000i 0.70711 + 0.70711i
    1.41421 + 0.00000i 0.70711 + 0.70711i

    If you want to derive the whole expression, I would investigate the
    meaning of the parameters instead of relying on pure algebra. For
    instance, the relation between Z and S parameters given in the wikipedia
    (http://en.wikipedia.org/wiki/Impedance_parameters),

    Z=sqrt(z)*(1+S)(1-S)^(-1)*sqrt(z)

    is just the vectorial generalization of the scalar equation

    Z=Z0*(1+ro)/(1-ro)

    I know that this is not the equation you are looking for, but the idea
    may be a starting point.

    Pere
     
  3. Thank you, gentlemen one and all, for your kind and helpful responses. I am evaluating your advice, as time permits, and hope to upload a pdf file of the excerpt from Abrie, 2009, sec 1.5 to Dropbox soon.

    Kind regards,

    Julian Grodzicky
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-