Counting polyominoes: a parallel implementation for cluster computing
ICCS'03 Proceedings of the 2003 international conference on Computational science: PartIII
| Hi-index | 0.00 |
Recently I. Jensen published a novel transfer-matrix algorithm for computing the number of polyominoes in a rectangular lattice. However, his estimation of the computational complexity of the algorithm (O((2)^n), where n is the size of the polyominoes), was based only on empirical evidence. In contrast, our research provides some solid proof. Our result is based primarily on an analysis of the number of some class of strings that plays a significant role in the algorithm. It turns out that this number is closely related to Motzkin numbers. We provide a rigorous computation that roughly confirms Jensen's estimation. We obtain the bound O(n^5^/^2(3)^n) on the running time of the algorithm, while the actual number of polyominoes is about C4.06^n/n, for some constant C0.