Statistical Information Propagation Across Operators for Dynamic Power Estimation on FPGAs

Yehya Nasser (1) , Jean-Christophe Prévotet (1) , Maryline Hélard (1) , J Lorandel (1)

1 IETR - Institut d'Électronique et des Technologies du numéRique

Performance, cost and energy consumption are the main key features to evaluate any embedded system. Low power system is needed then power optimization process is required. To optimize the power dissipation of an embedded system we have to estimate the power consumption first. This paper presents a new power estimation approach at early design phases which is based on the decomposition of a digital system into basic operators. Each operator has its own model which estimates the switching activity and the power consumption. By interconnecting several operators, statistical information like switching activities and percentage of logic high is then propagated to enable a global power estimation of a given system. The approach consists in simply adding the individual power consumption of each operator. The methodology has been evaluated in a use-case. The preliminary results indicate a promising speedup of the design process with less than 8.0% of error comparing to the classical power estimation tools.

Domaines: Sciences de l'ingénieur [physics]

Yehya Nasser, Jean-Christophe Prévotet, Maryline Hélard, J Lorandel. Statistical Information Propagation Across Operators for Dynamic Power Estimation on FPGAs . 12ème Colloque du GDR SoC/SiP, A Bordeaux du 14 au 16 Juin 2017, Jun 2017, Bordeaux, France. ⟨hal-01567196⟩