0
$\begingroup$

I'm trying to construct the transfer matrix for an arbitrary system with 1 input and 2 outputs, like a splitter (shown below) or 1x2 multi-mode interferometer. (Image borrowed from http://www.fiberstore.com/images/ckfinder/images/tutorial/1x2_Splitter.jpg)

a busy cat

Currently, I have the entire scattering matrix for my device:

$$ \begin{pmatrix} A_\text{in}^{(-)}\\ A_\text{out1}^{(+)}\\ A_\text{out2}^{(+)} \end{pmatrix} = \begin{pmatrix} S_{00} & S_{01} & S_{02}\\ S_{10} & S_{11} & S_{12}\\ S_{20} & S_{21} & S_{22}\\ \end{pmatrix} \begin{pmatrix} A_\text{in}^{(+)}\\ A_\text{out1}^{(-)}\\ A_\text{out2}^{(-)} \end{pmatrix} $$

where the $(+)$ and $(-)$ correspond to forward and backward propagating signals, respectively. I am looking for a transfer-matrix $\hat{T}$ such that $$ \begin{pmatrix} A_\text{out1}^{(+)}\\ A_\text{out1}^{(-)}\\ A_\text{out2}^{(+)}\\ A_\text{out2}^{(-)}\\ \end{pmatrix} = \begin{pmatrix} T_{00} & T_{01}\\ T_{10} & T_{11}\\ T_{20} & T_{21}\\ T_{30} & T_{31}\\ \end{pmatrix} \begin{pmatrix} A_\text{in}^{(+)}\\ A_\text{in}^{(-)}\\ \end{pmatrix} $$

Likewise, I am interested in the inverse transfer matrix such that $$ \begin{pmatrix} A_\text{in}^{(+)}\\ A_\text{in}^{(-)}\\ \end{pmatrix} = \begin{pmatrix} T_{00} & T_{01} & T_{02} & T_{03}\\ T_{10} & T_{11} & T_{12} & T_{13}\\ \end{pmatrix} \begin{pmatrix} A_\text{out1}^{(+)}\\ A_\text{out1}^{(-)}\\ A_\text{out2}^{(+)}\\ A_\text{out2}^{(-)}\\ \end{pmatrix} $$ which I figured I could just get from the pseudo-inverse of the previous result. Would this work? In both cases I am having trouble finding $\hat{T}$ because I am moving from a scattering matrix with 3 variables and 3 equations to either (1) an overdetermined system with 2 variables and 4 equations or (2) an underdetermined system with 4 variables and 2 equations.

It seems that generating a transfer-matrix for such a splitter should be quite straightforward, especially since I already have the scattering matrix data. Am I missing something here?

$\endgroup$

1 Answer 1

0
$\begingroup$

So I think I figured out what my problem is. After some reading transfer matrices are just convenient rearrangements of the same information contained within a scattering matrix such that you can easily cascade the matrices and make the math a bit nicer. However, for my purposes I can just write out the separate equations resulting from the 3x3 scattering matrices and solve the system brute-force. Now I'm just evaluating all my cascaded components (with equal i/o's) via transfer matrices then manually linking them up to the splitters.

TLDR; don't need the transfer matrices, scattering matrices do just fine on their own.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.