Boundary Constraints¶

Boundary constraints specify that at the start or at the end of the computation an indicated register has a given value. We define boundary constraints as a set \(B\) of triple \((r, c, e) \in \lbrace0,..,T \rbrace \times \lbrace 0,...,w-1 \rbrace \times F\), which means that the value of row \(r\) and column \(c\) is \(e\) in trace.