Convex programming upper bounds on the capacity of 2-D constraints

The capacity of 1-D constraints is given by the entropy of a corresponding stationary maxentropic Markov chain. Namely, the entropy is maximized over a set of probability distributions, which is defined by some linear equalities and inequalities. In this paper, certain aspects of this characterization are extended to 2-D constraints. The result is a method for calculating an upper bound on the capacity of 2-D constraints. The key steps are as follows: The maxentropic stationary probability distribution on square configurations is considered; set of linear equalities and inequalities is derived from this stationarity; the result is then a convex program, which can be easily solved numerically. Our method improves upon previous upper bounds for the capacity of the 2-D "no isolated bits" constraint, as well as certain 2-D RLL constraints.