Complexity theory for splicing systems

Abstract

This paper proposes a notion of time complexity in splicing systems and presents fundamental properties of SPLTIME, the class of languages with splicing system time complexity t(n). Its relations to classes based on standard computational models are explored. It is shown that for any function t(n), SPLTIME[t(n)] is included in 1-NSPACE[t(n)]. Expanding on this result, 1-NSPACE[t(n)] is characterized in terms of splicing systems: it is the class of languages accepted by a t(n)-space uniform family of extended splicing systems having production time O(t(n)) with regular rules described by finite automata with at most a constant number of states. As to lower bounds, it is shown that for all functions t(n) > log n, all languages accepted by a pushdown automaton with maximal stack height t(|x|) for a word x are in SPLTIME[t(n)]. From this result, it follows that the regular languages are in SPLTIME[O(log(n))] and that the context-free languages are in SPLTIME[O(n)]. © Springer-Verlag Berlin Heidelberg 2007.