Residue Arithmetic for Variation-Tolerant Design of Multiply-Add Units

This paper investigates the residue arithmetic as a solution for the design of variation-tolerant circuits. Motivated by the modular organization of residue processors, we comparatively study the sensitivity of residue arithmetic-based and binary processors to delay variations, and in particular the impact of delay variations onto the maximum critical path. Experiments are performed on two multiply-add (MAC) circuits based on residue and binary arithmetic. Results reveal that residue arithmetic-based circuits are up to 94% less sensitive to delay variation than binary circuits, thus leading to increased timing yield.

[1]  Ramakrishnan Venkatraman,et al.  Optimization strategies to improve statistical timing , 2009, 2009 10th International Symposium on Quality Electronic Design.

[2]  Kwang-Ting Cheng,et al.  Path selection for delay testing of deep sub-micron devices using statistical performance sensitivity analysis , 2000, Proceedings 18th IEEE VLSI Test Symposium.

[3]  Mohab Anis,et al.  On statistical timing analysis with inter- and intra-die variations , 2005, Design, Automation and Test in Europe.

[4]  Gian Carlo Cardarilli,et al.  Tradeoffs between residue number system and traditional FIR filters , 2001, ISCAS 2001. The 2001 IEEE International Symposium on Circuits and Systems (Cat. No.01CH37196).

[5]  Noel Menezes,et al.  Statistical timing analysis based on a timing yield model , 2004, Proceedings. 41st Design Automation Conference, 2004..

[6]  Haridimos T. Vergos,et al.  Modulo 2n±1 Adder Design Using Select-Prefix Blocks , 2003, IEEE Trans. Computers.

[7]  Keh-Jeng Chang,et al.  Modeling Sub-90nm On-Chip Variation Using Monte Carlo Method for DFM , 2007, 2007 Asia and South Pacific Design Automation Conference.

[8]  Ahmad A. Hiasat,et al.  High-Speed and Reduced-Area Modular Adder Structures for RNS , 2002, IEEE Trans. Computers.

[9]  Long-Ching Yeh,et al.  An Exploratory Study on Statistical Timing Analysis and Parametric Yield Optimization , 2007, 8th International Symposium on Quality Electronic Design (ISQED'07).

[10]  Nikolay Rubanov,et al.  An information theoretic framework to compute the MAX/MIN operations in parameterized statistical timing analysis , 2009, 2009 10th International Symposium on Quality Electronic Design.

[11]  David Blaauw,et al.  Statistical Analysis and Optimization for VLSI: Timing and Power , 2005, Series on Integrated Circuits and Systems.

[12]  Sani R. Nassif,et al.  Design for Manufacturability and Statistical Design - A Constructive Approach , 2007, Series on integrated circuits and systems.

[13]  Vassilis Paliouras,et al.  Mixed Radix-2 and High-Radix RNS Bases for Low-Power Multiplication , 2009, PATMOS.

[14]  Rob A. Rutenbar,et al.  Practical, fast Monte Carlo statistical static timing analysis: why and how , 2008, ICCAD 2008.

[16]  Michael A. Soderstrand,et al.  Residue number system arithmetic: modern applications in digital signal processing , 1986 .

[17]  Sani R. Nassif,et al.  Design for Manufacturability and Statistical Design: A Comprehensive Approach , 2006 .

[18]  Gian Carlo Cardarilli,et al.  Low-power adaptive filter based on RNS components , 2007, 2007 IEEE International Symposium on Circuits and Systems.

[19]  M. Anis,et al.  On efficient Monte Carlo-based statistical static timing analysis of digital circuits , 2008, ICCAD 2008.

[20]  James D. Meindl,et al.  Impact of die-to-die and within-die parameter fluctuations on the maximum clock frequency distribution for gigascale integration , 2002, IEEE J. Solid State Circuits.

[21]  Reto Zimmermann,et al.  Efficient VLSI implementation of modulo (2/sup n//spl plusmn/1) addition and multiplication , 1999, Proceedings 14th IEEE Symposium on Computer Arithmetic (Cat. No.99CB36336).

[22]  Giorgos Dimitrakopoulos,et al.  Efficient diminished-1 modulo 2/sup n/ + 1 multipliers , 2005, IEEE Transactions on Computers.

[23]  Farid N. Najm.  On the need for statistical timing analysis , 2005, Proceedings. 42nd Design Automation Conference, 2005..

[24]  G. Jullien,et al.  An algorithm for multiplication modulo (2/spl and/N-1) , 1996, Proceedings of the 39th Midwest Symposium on Circuits and Systems.