Exploitation of Flexibility in Sequential Networks

Abstract

We have seen how language equations can be solved by manipulating automata or FSMs, and that a largest solution can be obtained in terms of a deterministic automaton. If the solution is required as a finite state machine, then the solution can be made prefix-closed and (input)-progressive. At this point, we have the largest (non-deterministic, in general) FSM solution which contains all possible deterministic FSMs that are a solution; usually a deterministic FSM solution can be obtained by selecting a submachine of the largest solution. In some problems, it is enough to know that a solution exists and any one solution will be adequate. In other problems, we want a solution which can be implemented in an efficient way, say with small area, or power or delay.