V. Radulescu (Romania)
Modeling, hardware automata, decomposition, genetic algorithms.
Hardware automata are often very complex, which makes the design process very difficult. The decomposition of an automaton into a set of simpler subautomata leads to an easier manipulation of these components. While the full decomposition of automata has some advantages over the functional decomposition, the efficient identification of the subautomata is still an open problem. As deterministic algorithms have exponential complexity, the use of heuristic techniques may be the answer. The present paper focuses on the application of genetic algorithms on the full decomposition of automata, which exhibits very good results even for structures of high complexity.
Important Links:
Go Back
