A structure-oriented power modeling technique for macrocells

To characterize the power consumption of a macrocell, a general method involves recording the power consumption of all possible input transition events in the look-up tables. However, though this approach is accurate, the size of the table becomes very large. In this paper, we propose a new power modeling technique that takes advantage of the structural information of a macrocell. In this approach, a subset of primary inputs and internal nodes in the macrocell are selected as the state variables to build a state transition graph (STG). These state variables can model the steady-state transitions completely. Moreover, by selecting the characterization patterns properly, the STG can also model the glitch power in the macrocell accurately. To further simplify the complexity of the STG, an incomplete power modeling technique is presented. Without losing much accuracy, the property of compatible patterns is exploited for a macrocell to further reduce the number of edges in the corresponding STG. Experimental results show that our modeling techniques can provide SPICE-like accuracy, while the size of the look-up table is significantly reduced.