Efficient scheduling for imprecise timing based on fuzzy theory

In this paper, we propose a framework for design exploration in architectural synthesis which takes imprecise timing information into account. Based on the fuzzy theory, we designed a polynomial-time scheduling algorithm called inclusive scheduling which can efficiently construct schedules close to the ones obtained by exhaustive search. A heuristic for evaluating an imprecise schedule latency with respect to a latency constraint with multiple acceptability degrees is also proposed. Experimental results shows the effectiveness and efficiency of our approach by comparing designs generated by our algorithm with the traditional scheduling scheme assuming worst case (or typical case) timing values, as well as exhaustive method