Computer Science – Information Theory
Scientific paper
2009-03-04
Computer Science
Information Theory
submitted to IEEE Transactions on Information Theory
Scientific paper
The Coxeter lattices, which we denote $A_{n/m}$, are a family of lattices containing many of the important lattices in low dimensions. This includes $A_n$, $E_7$, $E_8$ and their duals $A_n^*$, $E_7^*$ and $E_8^*$. We consider the problem of finding a nearest point in a Coxeter lattice. We describe two new algorithms, one with worst case arithmetic complexity $O(n\log{n})$ and the other with worst case complexity O(n) where $n$ is the dimension of the lattice. We show that for the particular lattices $A_n$ and $A_n^*$ the algorithms reduce to simple nearest point algorithms that already exist in the literature.
Clarkson Vaughan I. L.
McKilliam Robby G.
Smith Warren D.
No associations
LandOfFree
Linear-time nearest point algorithms for Coxeter lattices does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Linear-time nearest point algorithms for Coxeter lattices, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Linear-time nearest point algorithms for Coxeter lattices will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-368458