Abstract
A new type of nonhomogeneous real time trellis automata, the so-called modular trellis automata, is introduced and various results concerning their normal forms, power, simulations, and decision problems are shown. Modular trellis automata are a mathematical abstraction, in the form of a recognizer, of an intuitive notion of an array of simple processors assembled in a simple and modular way. Distribution of processors in a real-time trellis automaton forms a two-dimensional structure called trellis. Basic characterizations and properties of modular trellises are summarized and modularity of various special trellises – regular, product, self-embedding, and self-overlapping – is investigated.
Get full access to this article
View all access options for this article.