Fast power estimation of large circuits

Our new technique for estimating transition probabilities of internal signals in combinational circuits uses Markov chains and reconvergence regions. To efficiently implement the computation, we use ROBDDs (reduced, ordered binary decision diagrams). Accounting for temporal dependence of signals, multiple concurrent transitions, and mutual dependence of internal signals, the technique provides an exact computation for small circuits and an approximate estimate for large circuits. Experimental results show the estimation technique is fast with only small inaccuracies.     «

Our new technique for estimating transition probabilities of internal signals in combinational circuits uses Markov chains and reconvergence regions. To efficiently implement the computation, we use ROBDDs (reduced, ordered binary decision diagrams). Accounting for temporal dependence of signals, multiple concurrent transitions, and mutual dependence of internal signals, the technique provides an exact computation for small circuits and an approximate estimate for large circuits. Experimental re...     »