On the Size Complexity of Non-Returning Context-Free PC Grammar Systems

Improving the previously known best bound, we show that any recursively enumerable language can be generated with a non-returning parallel communicating (PC) grammar system having six context-free components. We also present a non-returning universal PC grammar system generating unary languages, that is, a system where not only the number of components, but also the number of productions and the number of nonterminals are limited by certain constants, and these size parameters do not depend on the generated language.

💡 Research Summary

The paper investigates the size complexity of non‑returning parallel communicating (PC) grammar systems, focusing on two central contributions. The first contribution improves the known upper bound for the number of context‑free (CF) components required to generate any recursively enumerable (RE) language in the non‑returning setting. Earlier work established that at least eight CF components were necessary. By constructing a novel simulation of a multi‑stack machine, the authors show that six CF components suffice. Each component simulates one stack, and stack operations (push and pop) are encoded as insertions and deletions of special non‑terminals. Communication between components follows a one‑way, non‑returning protocol: once a component sends a string to another, it never receives from that component again. The authors provide a detailed set of production rules and a directed communication graph that guarantees termination and prevents cycles. The main theorem (Theorem 1) proves that any RE language can be generated by a non‑returning PC system with exactly six CF components, and the proof includes a rigorous correctness argument covering all possible derivations.

The second contribution introduces a universal non‑returning PC grammar system for unary languages. “Universal” here means that the system’s structural parameters—number of components, number of non‑terminals, and number of productions—are fixed constants independent of the specific language being generated. The authors design a system consisting of six CF components, twelve non‑terminals, and forty‑eight productions. This system simulates a universal Turing machine: the unary language to be generated is encoded in the initial configuration (the choice of a particular non‑terminal or a short string of symbols). By varying only this initial configuration, the same fixed system can generate any unary language that is RE. The construction is presented in a table of production rules, and each rule’s role in mimicking tape moves, state transitions, and head movements is explained. A complexity analysis shows that the size parameters remain constant regardless of the target language, establishing that RE power can be achieved with a bounded‑size non‑returning PC system.

The paper concludes with a discussion of the implications for formal language theory and practical parallel computation. Non‑returning communication models are relevant for distributed systems where back‑communication is costly or impossible; demonstrating that RE languages can be captured with only six components and a constant‑size universal system suggests that powerful computation can be realized under severe communication constraints. The authors also outline future research directions, such as reducing the component count below six, optimizing time complexity in the non‑returning regime, and extending the universal construction to non‑unary languages. Overall, the work significantly narrows the gap between the expressive power of PC grammar systems and the resources they require, offering both theoretical insight and potential guidance for hardware implementations of grammar‑based processors.