Trading Accuracy for Power with an Underdesigned Multiplier Architecture
Top Cited Papers
- 1 January 2011
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 346-351
- https://doi.org/10.1109/vlsid.2011.51
Abstract
We propose a novel multiplier architecture with tunable error characteristics, that leverages a modified inaccurate 2×2 building block. Our inaccurate multipliers achieve an average power saving of 31.78% - 45.4% over corresponding accurate multiplier designs, for an average error of 1.39% - 3.32%. Using image filtering and JPEG compression as sample applications we show that our architecture can achieve 2X - 8X better Signal-Noise-Ratio (SNR) for the same power savings when compared to recent voltage over-scaling based power-error tradeoff methods. We project the multiplier power savings to bigger designs highlighting the fact that the benefits are strongly design-dependent. We compare this circuit-centric approach to power-quality tradeoffs with a pure software adaptation approach for a JPEG example. We also enhance the design to allow for correct operation of the multiplier using a residual adder, for non error-resilient applications.Keywords
This publication has 12 references indexed in Scilit:
- Energy-aware probabilistic multiplierPublished by Association for Computing Machinery (ACM) ,2009
- Design Methodology for Low Power and Parametric Robustness Through Output-Quality Modulation: Application to Color-Interpolation FilteringIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2009
- Energy Aware Computing through Probabilistic Switching: A Study of LimitsIEEE Transactions on Computers, 2005
- Discrete Fourier transform using summation by partsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Arithmetic Data Value SpeculationLecture Notes in Computer Science, 2005
- Speeding up processing with approximation circuitsComputer, 2004
- Low power parallel multiplier design for DSP applications through coefficient optimizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Reducing power by optimizing the necessary precision/range of floating-point arithmeticIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2000
- Energy-efficient signal processing via algorithmic noise-tolerancePublished by Association for Computing Machinery (ACM) ,1999
- Fourier analysis and signal processing by use of the Mobius inversion formulaIEEE Transactions on Acoustics, Speech, and Signal Processing, 1990