As leakage power and total power is a more and more dramatic issue in very deep submicron technologies, this paper explores new design methodologies for designing leakage tolerant digital architectures, based on architectural parameters like activity, logical depth, number of transitions for achieving a given task and total number of gates. Various architectures for a same logic function are compared at very low Vdd and Vth that define the optimal total power consumption of each architecture. The first proposed design method selects the best architecture out of a set of architectures (baseline, sequential, parallel, pipelined, etc..) at optimal Vdd and threshold voltages Vth, while a second design method takes Vdd and threshold voltages Vth as given constraints. © Springer-Verlag Berlin Heidelberg 2006.
CITATION STYLE
Piguet, C., Schuster, C., & Nagel, J. L. (2006). Static and dynamic power reduction by architecture selection. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4148 LNCS, pp. 659–668). Springer Verlag. https://doi.org/10.1007/11847083_65
Mendeley helps you to discover research relevant for your work.