arXiv++ Combinatorics

The subword complexity of a word $w$ over a finite alphabet $\mathcal{A}$ is a function that assigns for each positive integer $n$, the number of distinct subwords of length $n$ in $w$. The subword complexity of a word is a good measure of the randomness of the word and gives insight to what the word itself looks like. In this paper, we discuss the properties of subword complexity sequences, and consider different variables that influence their shape. We also compute the number of distinct subword complexity sequences for certain lengths of words over different alphabets, and state some conjectures about the growth of these numbers.

In this article, we study subword complexity of colorings of regular trees. We characterize colorings of bounded subword complexity and study Sturmian colorings, which are colorings of minimal unbounded subword complexity. We classify Sturmian colorings using their type sets. We show that any Sturmian coloring is a lifting of a coloring on a quotient graph of the tree which is a geodesic or a ray with loops possibly attached, thus a lifting of an "infinte word". We further give a complete characterization of the quotient graph for eventually periodic ones.

We investigate the computational power of periodically iterated morphisms, also known as D0L systems with periodic control, PD0L systems for short. These systems give rise to a class of one-sided infinite sequences, called PD0L words. We construct a PD0L word with exponential subword complexity, thereby answering a question raised by Lepisto (1993) on the existence of such words. We solve another open problem concerning the decidability of the first-order theories of PD0L words; we show it is already undecidable whether a certain letter occurs in a PD0L word. This stands in sharp contrast to the situation for D0L words (purely morphic words), which are known to have at most quadratic subword complexity, and for which the monadic theory is decidable. The main result of our paper, leading to these answers, is that every computable word w over an alphabet Sigma can be embedded in a PD0L word u over an extended alphabet Gamma in the following two ways: (i) such that every finite prefix of w is a subword of u, and (ii) such that w is obtained from u by erasing all letters from Gamma not in Sigma. The PD0L system generating such a word u is constructed by encoding a Fractran program that computes the word w; Fractran is a programming language as powerful as Turing Machines. As a consequence of (ii), if we allow the application of finite state transducers to PD0L words, we obtain the set of all computable words. Thus the set of PD0L words is not closed under finite state transduction, whereas the set of D0L words is. It moreover follows that equality of PD0L words (given by their PD0L system) is undecidable. Finally, we show that if erasing morphisms are admitted, then the question of productivity becomes undecidable, that is, the question whether a given PD0L system defines an infinite word.

We show that the subword complexity function p_x(n), which counts the number of distinct factors of length n of a sequence x, is k-synchronized in the sense of Carpi if x is k-automatic. As an application, we generalize recent results of Goldstein. We give analogous results for the number of distinct factors of length n that are primitive words or powers. In contrast, we show that the function that counts the number of unbordered factors of length n is not necessarily k-synchronized for k-automatic sequences.

We derive a lower bound for the subword complexity of the base-$b$ expansion ($b\geq 2$) of all real numbers whose irrationality exponent is equal to 2. This provides a generalization of a theorem due to Ferenczi and Mauduit. As a consequence, we obtain the first lower bound for the subword complexity of the number $e$ and of some other transcendental exponential periods.

Let $γ_{a,b}(n)$ be the number of smooth words of length $n$ over the alphabet $\{a,b\}$ with $a<b$. Say that a smooth word $w$ is \emph{left fully extendable} (LFE) if both $aw$ and $bw$ are smooth. In this paper, we prove that for any positive number $ξ$ and positive integer $n_{0}$ such that the proportion of $b$'s is larger than $ξ$ for each LFE word of length exceeding $n_0$, there are two constants $c_{1}\,\textrm{and}\, c_{2}$ such that for each positive integer $n$, one has {eqnarray} c_{1}\cdot n^{\frac{\log (2b-1)}{\log (1+(a+b-2)(1-ξ))}}<γ_{a,b}(n)< c_2\cdot n^{\frac{\log (2b-1)}{\log (1+(a+b-2)ξ)}}. {eqnarray} In particular, taking $a=1\text{and}b=2$ in the above inequalities arrives at Huang and Weakley's result. Moreover, for 2-letter even alphabet $\{a,b\}$, there are two suitable constants $c_1,\,c_2$ such that \{eqnarray} c_{1}\cdot n^{\frac{\log (2b-1)}{\log ((a+b)/2)}}<γ_{a,b}(n)< c_2\cdot n^{\frac{\log (2b-1)}{\log ((a+b)/2)}}\textit{for each positive integer $n$}.\{eqnarray}

We consider a new family of factorial languages whose subword complexity grows as $Θ(n^α)$, where $α$ is the root of some transcendent equation. Analytical methods and in particular, a corollary of the Wiener-Pitt theorem, are used to find the asymptotic growth of the complexity. Factorial languages considered are languages of arithmetical factors of some Toeplitz words. So, we describe a new family of words with an unusual growth of arithmetical complexity.

Decimal expansions of classical constants such as $\sqrt2$, $π$ and $ζ(3)$ have long been a source of difficult questions. In the case of Laurent series with coefficients in a finite field, where no carry-over difficulties appear, the situation seems to be simplified and drastically different. On the other hand, Carlitz introduced analogs of real numbers such as $π$, $e$ or $ζ(3)$. Hence, it became reasonable to enquire how "complex" the Laurent representation of these "numbers" is. In this paper we prove that the inverse of Carlitz's analog of $π$, $Π_q$, has in general a linear complexity, except in the case $q=2$, when the complexity is quadratic. In particular, this implies the transcendence of $Π_2$ over $\F_2(T)$. In the second part, we consider the classes of Laurent series of at most polynomial complexity and of zero entropy. We show that these satisfy some nice closure properties.

In this paper we undertake the general study of the Abelian complexity of an infinite word on a finite alphabet. We investigate both similarities and differences between the Abelian complexity and the usual subword complexity. While the Thue-Morse minimal subshift is neither characterized by its Abelian complexity nor by its subword complexity alone, we show that the subshift is completely characterized by the two complexity functions together. We give an affirmative answer to an old question of G. Rauzy by exhibiting a class of words whose Abelian complexity is everywhere equal to 3. We also investigate links between Abelian complexity and the existence of Abelian powers. Using van der Waerden's Theorem, we show that any minimal subshift having bounded Abelian complexity contains Abelian k-powers for every positive integer k. In the case of Sturmian words we prove something stronger: For every Sturmian word w and positive integer k, each sufficiently long factor of w begins in an Abelian k-power.

In this paper we explore various interconnections between rich words, Sturmian words, and trapezoidal words. Rich words, first introduced in arXiv:0801.1656 by the second and third authors together with J. Justin and S. Widmer, constitute a new class of finite and infinite words characterized by having the maximal number of palindromic factors. Every finite Sturmian word is rich, but not conversely. Trapezoidal words were first introduced by the first author in studying the behavior of the subword complexity of finite Sturmian words. Unfortunately this property does not characterize finite Sturmian words. In this note we show that the only trapezoidal palindromes are Sturmian. More generally we show that Sturmian palindromes can be characterized either in terms of their subword complexity (the trapezoidal property) or in terms of their palindromic complexity. We also obtain a similar characterization of rich palindromes in terms of a relation between palindromic complexity and subword complexity.

We study matrices of morphisms preserving the family of words coding 3-interval exchange transformations. It is well known that matrices of morphisms preserving sturmian words (i.e. words coding 2-interval exchange transformations with the maximal possible factor complexity) form the monoid $\{\boldsymbol{M}\in\mathbb{N}^{2\times 2} | \det\boldsymbol{M}=\pm1\} = \{\boldsymbol{M}\in\mathbb{N}^{2\times 2} | \boldsymbol{M}\boldsymbol{E}\boldsymbol{M}^T = \pm\boldsymbol{E}\}$, where $\boldsymbol{E} = (\begin{smallmatrix}0&1 -1&0\end{smallmatrix})$. We prove that in case of exchange of three intervals, the matrices preserving words coding these transformations and having the maximal possible subword complexity belong to the monoid $\{\boldsymbol{M}\in\mathbb{N}^{3\times 3} | \boldsymbol{M}\boldsymbol{E}\boldsymbol{M}^T = \pm\boldsymbol{E},\ \det\boldsymbol{M}=\pm 1\}$, where $\boldsymbol{E} = \Big(\begin{smallmatrix}0&1&1 -1&0&1 -1&-1&0\end{smallmatrix}\Big)$.

Let $A_q$ be a $q$-letter alphabet and $w$ be a right infinite word on this alphabet. A subword of $w$ is a block of consecutive letters of $w$. The subword complexity function of $w$ assigns to each positive integer $n$ the number $f_w(n)$ of distinct subwords of length $n$ of $w$. The gap function of an infinite word over the binary alphabet $\{0,1 \}$ gives the distances between consecutive 1's in this word. In this paper we study infinite binary words whose gap function is injective or "almost injective". A method for computing the subword complexity of such words is given. A necessary and sufficient condition for a function to be the subword complexity function of a binary word whose gap function is strictly increasing is obtained.