A 0.25V 460nW Asynchronous Neural Signal Processor with Inherent Leakage Suppression
Journal
IEEE Journal of Solid-State Circuits
Journal Volume
48
Journal Issue
4
Pages
897-906
Date Issued
2013-04
Author(s)
Abstract
Further power and energy reductions via technology and voltage scaling have become extremely difficult due to leakage and variability issues. In this paper, we present a robust and energy-efficient computation architecture exploiting an asynchronous timing strategy to dynamically minimize leakage and to self-adapt to process variations and different operating conditions. Based on a logic topology with built-in leakage suppression, the prototype asynchronous neural signal processor demonstrates robust sub-threshold operation down to 0.25 V, while consuming only 460 nW in 0.03 mm2 in a 65 nm CMOS technology. These results represent a 4.4× reduction in power, a 3.7× reduction in energy and a 2.2$\times$ reduction in power density, when compared to the state-of-the-art processors. © 1966-2012 IEEE.
SDGs
Type
journal article