Computational power of insertion-deletion (P) systems with rules of size two

This article investigates insertion-deletion systems of small size, where at most two symbols can be used in the description of insertion or deletion rules in a context-free or contextual manner. The basic result shows a characterization by context-free grammars of insertion-deletion systems, which insert or delete one symbol in one symbol left context (systems of size (1, 1, 0; 1, 1, 0)). If context-free insertion or deletion rules are considered (systems of size (2, 0, 0; 1, 1, 0) or (1, 1, 0; 2, 0, 0)), then we show that corresponding systems are not computationally complete. However, if the insertion and the deletion operations having same size as above are considered in the distributed framework of P systems, then the computational power strictly increases and the obtained models become computationally complete. The article also shows that if context-free insertion and deletion rules of two symbols (of size (2, 0, 0; 2, 0, 0)) are used in combination with P systems, then the obtained model is still not computationally complete. Finally some open problems are presented.