In this paper, we present biconditional binary decision diagrams (BBDDs), a novel canonical representation form for Boolean functions. BBDDs are binary decision diagrams where the branching condition, and its associated logic expansion, is biconditional on two variables. Empowered by reduction and ordering rules, BBDDs are remarkably compact and unique for a Boolean function. The interest of such representation form in modern electronic design automation (EDA) is twofold. On the one hand, BBDDs improve the efficiency of traditional EDA tasks based on decision diagrams, especially for arithmetic intensive designs. On the other hand, BBDDs represent the natural and native design abstraction for emerging technologies where the circuit primitive is a comparator, rather than a simple switch. We provide, in this paper, a solid ground for BBDDs by studying their underlying theory and manipulation properties. Thanks to an efficient BBDD software package implementation, we validate 1) speed-up in traditional decision diagrams applications with up to 4.4× gain with respect to other DDs, and 2) improved synthesis of circuits in emerging technologies, with about 32% shorter critical path than state-of-art synthesis techniques.
CITATION STYLE
Amaru, L., Gaillardon, P. E., & De Micheli, G. (2014). Biconditional binary decision diagrams: A novel canonical logic representation form. IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 4(4), 487–500. https://doi.org/10.1109/JETCAS.2014.2361058
Mendeley helps you to discover research relevant for your work.