Energy-Efficient Hybrid Analog/Digital Approximate Computation in Continuous Time
Authors:
Top Cited Papers
- 29 April 2016
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 51 (7) , 1514-1524
Abstract
We present a unit that performs continuous-time hybrid approximate computation, in which both analog and digital signals are functions of continuous time. Our 65 nm CMOS prototype system is capable of solving nonlinear differential equations up to 4th order, and is scalable to higher orders. Nonlinear functions are generated by a programmable, clockless, continuoustime 8-bit hybrid architecture (ADC + SRAM + DAC). Digitally assisted calibration is used in all analog/mixed-signal blocks. Compared to the prior art, our chip makes possible arbitrary nonlinearities and achieves 16× lower power dissipation, thanks to technology scaling and extensive use of class-AB analog blocks. Typically, the unit achieves a computational accuracy of about 0.5% to 5% RMS, solution times from a fraction of 1 μs to several hundred μs, and total computational energy from a fraction of 1 nJ to hundreds of nJ, depending on equation details. Very significant advantages are observed in computational speed and energy (over two orders of magnitude and over one order of magnitude, respectively) compared to those obtained with a modern microcontroller for the same RMS error.Keywords
Funding Information
- National Science Foundation (CNS 1239134)
This publication has 12 references indexed in Scilit:
- Continuous-time hybrid computation with programmable nonlinearities2015
- Approximate computing: An emerging paradigm for energy-efficient design2013
- SleepWalker: A 25-MHz 0.4-V Sub-$\hbox{mm}^{2}$ 7-$\mu\hbox{W/MHz}$ Microcontroller in 65-nm LP/GP CMOS for Low-Carbon Wireless Sensor NodesIEEE Journal of Solid-State Circuits, 2012
- A Clockless ADC/DSP/DAC System with Activity-Dependent Power Dissipation and No AliasingIEEE International Solid-State Circuits Conference, 2008
- Ultra Low-Power Low-Voltage Integrated Preamplifier Using Class-AB Op-Amp for Biomedical Sensor Application2007
- Frequency-Dynamic Range-Power2006
- A VLSI analog computer/math co-processor for a digital computer2005
- Two-Dimensional Differential Equations1995
- The CMOS gain-boosting techniqueAnalog Integrated Circuits and Signal Processing, 1991
- The Role of JSC Engineering Simulation in the Apollo ProgramSIMULATION, 1991