Maximum-sized antichains in minimal posets (Q1081623): Difference between revisions

Let P be a poset and A(P) the set of its antichains. It is known that A(P) becomes a distributive lattice if for \(\alpha\),\(\beta\in A(P)\) we put \(\alpha\leq \beta\) iff for each \(a\in \alpha\) there exists \(b\in \beta\) such that \(a\leq b\) in P. Let \(A_ m(P)\) be the set of those antichains of P which are maximum-sized (i.e. maximal with respect to cardinality). It has been proved by Dilworth in 1960 that the set \(A_ m(P)\) forms a distributive lattice under the above ordering. Recently [Algebra Univers. 17, 73-86 (1983; Zbl 0524.06003)] the author has established the converse of this result, by showing that for every distributive lattice L there is always a poset Q such that \(L\cong A_ m(Q)\). This Q is not unique, rather, for every distributive lattice L, there exist infinitely many posets R such that \(L\cong A_ m(R)\). On the basis of this the author defines a poset Q to be minimal iff \(| Q| \leq | R|\) for every poset R with \(A_ m(R)\cong A_ m(Q)\), and in this paper he determines completely the structure of every minimal poset.