Horizontal Info Mixing Primitive in 2D Shifts

• Horizontal information mixing primitive is a finitely-checkable framework that employs horizontal transition matrices and connecting operators to verify full mixing in two-dimensional shifts.
• It uses reduction techniques and invariant cycle conditions to transform local forbidden patterns into robust global dynamical properties like topological mixing and strong specification.
• The method provides a finite-step, algorithmically implementable approach to establish complete mixing and specification in symbolic dynamical systems.

The horizontal information mixing primitive is a central concept developed for the systematic verification of mixing properties in two-dimensional shifts of finite type. It provides a finite, checkable framework for demonstrating that the dynamical system induced by local $2\times2$ constraints on a finite alphabet achieves full mixing in the horizontal direction. The approach rests on the construction of horizontal transition matrices, reduction techniques involving connecting operators, and combinatorial criteria for primitivity, which collectively yield a rigorous path from local forbidden patterns to global properties such as topological mixing and strong specification (Ban et al., 2011).

1. Definition and Construction of Horizontal Transition Matrix

For a shift of finite type over a finite alphabet $\mathcal S_p = \{0,1,\dots,p-1\}$ governed by a set $\mathcal B \subset \mathcal S_p^{2\times2}$ of forbidden $2\times2$ blocks, the horizontal information mixing primitive is formalized through the family of horizontal transition matrices $\{H_n\}$. Each column of height $n$ is regarded as a word $u\in \mathcal S_p^n$, indexed by the bijection

$\psi(u_1,\ldots,u_n)=1+\sum_{k=1}^{n}u_k p^{n-k},$

mapping columns to the integers $1,\dots,p^n$. The horizontal transition matrix $H_n$ is a $p^n\times p^n$ zero-one matrix whose $(i,j)$ entry records whether columns $u$ and $v$ (with indices $i$ and $j$) may stand side by side without creating a forbidden $2\times2$ block: $$(H_n)_{i,j} = \begin{cases} 1, & \text{if } (u_k,u_{k+1},v_k,v_{k+1})\notin\mathcal B \text{ for all } 1\leq k\leq n-1,\ 0, & \text{otherwise.} \end{cases}$$ For $n=2$, $H_2$ coincides with the classical $p^2\times p^2$ matrix recording admissible $2\times2$ blocks.

2. Primitivity Criteria

A nonnegative square matrix $A$ is primitive if $\exists\,K\geq1$ such that $A^K_{i,j}>0$ for all $i,j$. The horizontal mixing primitive is attained once $H_n$ is primitive for all $n\geq2$. Primitivity guarantees that sufficiently long horizontal concatenations of admissible columns can produce arbitrary admissible horizontal patterns, encoding the absence of global forbidden configurations.

3. Reduction via Connecting Operators

To enable inductive proofs and practical computations, the approach introduces for each $m\geq2$ a connecting operator $\mathbb S_{m}=[S_{m;\alpha,\beta}]$, with $(S_{m;\alpha,\beta})_{k,l}\in\{0,1\}$ indicating the admissibility of an $(m+1)\times2$ block linking codes $\alpha$, $\beta$ with interior states $(k,l)$. The essential reduction identities are

$$H_{m,n+1;\alpha;\beta}^{(k)} = \sum_{l=1}^{p^{m-1}} (S_{m;\alpha,\beta})_{k,l} H_{m,n;\beta}^{(l)},$$

and, for $q$-step extensions,

$$H_{m,n+q;\beta_1,\dots,\beta_{q+1}}^{(k)} = \sum_{l=1}^{p^{m-1}} (S_{m;\beta_1,\beta_2}\cdots S_{m;\beta_{q},\beta_{q+1}})_{k,l}\, H_{m,n;\beta_{q+1}}^{(l)},$$

enabling the reduction of high-order transition matrices to lower-order ones, which is fundamental for finite-step verification.

4. Finitely-Checkable Sufficient Conditions

Two combinatorial criteria for primitivity are provided:

A. $S$-Invariant Diagonal Cycles:

With the diagonal index set $\mathcal D_p = \{1+(p+1)j : 0\leq j<p\}$, a cycle $\overline\beta_q = \beta_1 \cdots \beta_q \beta_1$ using $\beta_i\in\mathcal D_p$ is $S$-invariant of order $(m,q)$ if there is $\mathcal K\subset \{1,\dots,p^{m-1}\}$ such that

$$\sum_{k\in\mathcal K} (S_{m;\beta_1,\beta_2}\cdots S_{m;\beta_q,\beta_1})_{k,\ell}\geq1\quad \forall\,\ell\in\mathcal K.$$

If $H_2$ is non-degenerated, the existence of such a cycle together with the primitivity, for $2\leq n\leq q+1$, of the aggregated blocks $\sum_{l\in\mathcal K} H_{m,n;\beta_1}^{(l)}$, ensures $H_n$ is primitive for all $n$.

B. Primitive Commutative Cycles:

Construct cycles $I_q = i_1\cdots i_q i_1$ and $J_{q'} = j_1\cdots j_{q'} j_1$ with $j_1=i_1$, consider their concatenations, and define a diagonal index $\bar\alpha$ and interior indices $K,L$. If $H_{m,2;\bar\alpha}^{(K)}$ and $H_{m,2;\bar\alpha}^{(L)}$ are primitive and the connecting condition $(S_{m;\bar\alpha,\bar\alpha})_{K,L}=1$ or $(S_{m;\bar\alpha,\bar\alpha})_{L,K}=1$ holds, this is an $H$-primitive commutative cycle pair. In the non-degenerate case, existence of such a pair suffices for primitivity of all $H_n$.

5. Finite-Step Verification Algorithm

A finite algorithm for establishing horizontal mixing primitive proceeds as follows:

• Compute $H_2$ and check for zero-free rows/columns.
• Compute connecting operators $S_m$ (and optionally $C_m$) for $2\leq m\leq M$.
• Search for $S$-invariant diagonal cycles; if found, check primitivity of $\sum_{l\in\mathcal K} H_{m,n;\beta_1}^{(l)}$ for $n=2,\dots,q+1$.
• If the cycle condition fails, search for short commutative cycle pairs and test the $H$-primitive condition.
• Success in either verification implies $H_n$ is primitive for all $n\geq2$.

6. Connection to Topological Mixing and Specification

Horizontal mixing primitive is a key ingredient for demonstrating mixing properties of the full two-dimensional shift $\Sigma(\mathcal B)$. When both horizontal $\{H_n\}$ and vertical families $\{V_n\}$ are primitive, and supplementary corner-extendability and crisscross-extendability conditions are met, Theorem 3.14 establishes that $\Sigma(\mathcal B)$ is topologically mixing. For strong specification, an additional hole-filling condition (HFC) is required, ensuring any annular admissible pattern of sufficient width is fillable at its center. Coupled with finite-step primitivity tests, this completes the path to strong specification (Ban et al., 2011).

7. Significance and Finiteness of the Mixing Test

The horizontal information mixing primitive provides a systematic and finitely-checkable route from local forbidden patterns to global dynamical properties of two-dimensional shifts of finite type. The framework relies exclusively on the analysis of transition matrices and their combinatorial reductions, offering a practical and rigorous mechanism for verifying intricate mixing behaviors and specification in symbolic dynamical systems. All procedural steps terminate in finitely many steps, making the primitive not only theoretically robust but also algorithmically implementable.