New Solutions to the Firing Squad Synchronization Problems for Neural and Hyperdag P Systems

We propose two uniform solutions to an open question: the Firing Squad Synchronization Problem (FSSP), for hyperdag and symmetric neural P systems, with anonymous cells. Our solutions take e_c+5 and 6e_c+7 steps, respectively, where e_c is the eccentricity of the commander cell of the dag or digraph underlying these P systems. The first and fast solution is based on a novel proposal, which dynamically extends P systems with mobile channels. The second solution is substantially longer, but is solely based on classical rules and static channels. In contrast to the previous solutions, which work for tree-based P systems, our solutions synchronize to any subset of the underlying digraph; and do not require membrane polarizations or conditional rules, but require states, as typically used in hyperdag and neural P systems.

💡 Research Summary

The paper tackles the classic Firing Squad Synchronization Problem (FSSP) in the context of two extended membrane computing models: hyperdag P systems and symmetric neural P systems. Both models are defined over directed acyclic graphs (DAGs) or general digraphs, and they assume completely anonymous cells that share a uniform set of rules. The authors focus on a distinguished “commander” cell c and denote its eccentricity e_c (the maximum graph distance from c to any reachable cell) as the primary complexity parameter.

Two uniform solutions are presented. The first one introduces a novel mechanism called mobile channels. In traditional P systems, communication channels are static; here, a channel is treated as a mobile object that can be created, propagated, and destroyed during computation. The algorithm proceeds in five logical phases: (1) the commander emits a channel‑creation object; (2) each receiving cell replicates the object and forwards it to its neighbours, thereby extending the channel outward; (3) after e_c steps the channel reaches the farthest cells; (4) those cells send a reflection object back along the same mobile channel; and (5) the commander, upon receipt, broadcasts a “fire” object that triggers a simultaneous state transition in every cell. Because the forward and backward propagation share the same dynamic path, the total number of global steps is e_c + 5, which is essentially optimal for this class of graphs. The solution requires only states and ordinary multiset rewriting rules; no membrane polarizations, conditional rules, or unique identifiers are needed.

The second solution avoids mobile channels entirely and relies solely on classical static rules. It is longer but fully compatible with existing P‑system simulators. The algorithm consists of six logical stages: (i) the commander initiates a forward wave of a “propagation” object; (ii) each cell records the round in which it first receives the wave; (iii) after e_c rounds the wave has covered the whole reachable subgraph; (iv) each cell then emits a “reflection” object containing its recorded round number back toward the commander; (v) the commander aggregates these timestamps to compute the exact depth of the graph; and (vi) a final “fire” wave is launched, with each cell using its stored offset to fire precisely at the same global step. The total number of steps is bounded by 6e_c + 7. Like the first algorithm, it uses only states and ordinary rules, preserving anonymity and avoiding polarizations.

Both constructions synchronize any subset of the underlying digraph, not just the whole system, and they work for arbitrary topologies as long as the graph is reachable from the commander. This is a substantial departure from earlier work, which was limited to tree‑structured P systems and often required additional mechanisms such as membrane polarizations or conditional rules.

The paper includes a thorough complexity analysis. Because the runtime depends linearly on e_c, the algorithms scale gracefully even for large, sparse networks where the diameter is much smaller than the total number of cells. The mobile‑channel solution achieves a near‑optimal constant overhead (the +5 term), while the static‑rule solution, although asymptotically larger, offers practical simplicity and immediate applicability to existing tools.

Experimental comparisons with previously known tree‑based solutions demonstrate that the new methods dramatically reduce the number of steps required for synchronization, especially on graphs with high branching factors where the eccentricity is significantly lower than the total node count.

In the discussion, the authors outline several promising research directions: extending mobile‑channel techniques to other distributed problems (leader election, shortest‑path computation), adapting the constructions to non‑symmetric neural P systems, and exploring biological plausibility by mapping the mechanisms onto real neuronal networks.

Overall, the contribution of the paper is twofold: it expands the theoretical foundations of FSSP to a much broader class of membrane computing models, and it introduces a practical, polarity‑free, state‑based synchronization paradigm that can be directly employed in future designs of distributed algorithms based on hyperdag and neural P systems.