Context-free sets of infinite words

11Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

In this paper we give some new results about context-free sets of infinite words. The presentation will be a generalization of McNaughton's approach in [7], where he analyzed regular sets of infinite words. However, our extension to the regular case is not straightforward and thus distinguishes from the approach given in [4]. Some of the results given below originate from two papers by Nivat [9,10], others are unpublished supplementary results due to Nivat and Boasson. We recall from [9] that to each context-free grammar G one can associate an operator Ĝ, which has a unique fixed point over finite words and a greatest fixed point over finite and infinite words, each of them being the vector of languages generated by the non-terminals of G. We then show that any context-free set of infinite words can be obtained by a substitution of some context-free languages into a regular set of infinite words. In the sequel the notions of adherence and center of context-free languages are introduced and analyzed to establish a link between the infinite words and the language generated by a grammar.

Cite

CITATION STYLE

APA

Boasson, L. (1979). Context-free sets of infinite words. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 67 LNCS, pp. 1–9). Springer Verlag. https://doi.org/10.1007/3-540-09118-1_1

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free