|Dec19-12, 01:37 AM||#1|
Hey all, I have been playing around with a special type of matrix and am wondering if anyone knows of some literature about it. I have been calling it a pseudo-orthogonal matrix but would like to learn if it has a real name or if we can come up with a better name. The characteristics of the matrix are as follows:
1) The matrix is composed of only ones and zeros
2) Each row and each column have the same number of ones in it. (If there are 3 ones in each row/column then I call a 3rd order matrix)
3) Between any two rows, there is one and only one common column with a one.
Here is an example of what I call a 3rd order pseudo-orthogonal matrix. Let's call him 'M'
1 1 1 0 0 0 0
1 0 0 1 1 0 0
1 0 0 0 0 1 1
0 1 0 1 0 1 0
0 1 0 0 1 0 1
0 0 1 1 0 0 1
0 0 1 0 1 1 0
I call it a pseudo inverse because inv(M) = M/2-1/6 , i.e. with adding and multiplying by constants I can arrive at the inverse of M.
Has anyone played with something like this before? I am hoping to gleen information to help me generate higher order matrices of this type.
|Dec19-12, 02:42 AM||#2|
Hey orthogonal and welcome to the forums.
I've never played with this kind of thing (I haven't had to): do you have a reason for doing this: (pure curiosity or do you have an application in mind)?
I don't know whether this would help but error correcting code matrices (in binary XOR) might share common properties with this (it's just a hunch and its probably wrong anyway, but you never know!).
|Dec19-12, 10:15 AM||#3|
I am working on the applications but I found the matrix by analyzing the matching card game Spot it! It is a card game which has 8 symbols per card with one and only one matching symbol between any two cards. If you follow the link above you can play a demo.
|Dec19-12, 05:57 PM||#4|
How about considering the eigenvalues to create a higher dimensional matrix (as well as the eigenvectors)?
|Dec30-12, 11:21 PM||#5|
I have some findings to report:
I have written a program which can generate up to order 6 successfully, but when my code attempts to do order 7 it chokes (24 hours + with no solution returned!).
After googling around some more I found a poster presentation which describes the problem using mutually orthogonal Latin squares. It looks like I have some reading to do to catch up on all this higher order geometry stuff.
|Similar Threads for: Pseudo-orthogonal matrix?|
|pseudo inverse matrix||Linear & Abstract Algebra||1|
|Proving transpose of orthogonal matrix orthogonal||Calculus & Beyond Homework||4|
|Orthogonal matrices prove: T is orthogonal iff [T]_bb is an orthogonal matrix||Calculus & Beyond Homework||0|
|[SOLVED] Pseudo orthogonal group||General Physics||204|
|[SOLVED] Pseudo orthogonal group||General Physics||8|