2
$\begingroup$

Given a prescribed trajectory, is it possible to construct an interval exchange having this trajectory?

For example, given a 3-letter word (like aaabbbccabcaaa ), is it possible to construct a 3- interval exchange with a point having this word as the beginning of its trajectory?

what necessary conditions on a given word to be the trarjectory of a IET can be found?

For the relation between coding and interval exchange, see e.g. : http://combinat.sagemath.org/doc/reference/combinat/sage/combinat/iet/tutorial.html#orbit-and-symbolic-coding

Improve this question
$\endgroup$

1 Answer 1

Reset to default
5
$\begingroup$

The complexity of an infinite sequence $x$ is a sequence $C(n)$, where $C(n)$ is the number of distinct blocks of length $n$ in $x$. For an interval exchange with $k$ symbols, it's not hard to show that $C(n)=(k-1)n+1$. If your word has more complexity than this, it can never appear as the coding sequence of an IET.

Improve this answer
$\endgroup$
2
  • $\begingroup$ where can i find the proof of this fact? $\endgroup$
    – user8991
    Commented Mar 17, 2014 at 5:51
  • 1
    $\begingroup$ If $E$ is the set of endpoints of intervals, then $|E|=k-1$. $E\cup T^{-1}E\cup \ldots\cup T^{-(n-1)}E$ has cardinality at most $n(k-1)$, and so the complement in [0,1] has at most $1+(k-1)n$ intervals. If two points are not separated by one of these endpoints, they have the same $n$-step coding. $\endgroup$
    – Anthony Quas
    Commented Mar 17, 2014 at 21:47

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .