Sign Bit Reduction Encoding For Low Power Applications

In this paper, a low power technique, called SBR (Sign Bit Reduction), which reduces the energy consumption in multipliers as well as data buses is proposed. The technique reduces the number of sign bits in the data transfer as well as in the multiplication process. This feature enables us to use the encoding technique for both the transfer of the data and its multiplication at the destination without any need for an intermediate decoding step. Simple circuits are used as the SBR decoder and encoder. The efficacy of the technique is evaluated for both voice and random data. The results of applying the voice data to a 16-bit multiplier implemented with this scheme shows energy consumption up to 11.4% compared to those of a 2’s complement implementation, while the number of required clock periods for the multiplication process is reduced up to 14.5%. The results of applying the SBR technique to a 30-tap FIR filter show up to 9.6% reduction in the energy consumption and up to 13.4% reduction in the required clock cycles. Finally, for voice data and random inputs, the use of the technique for a 16-bit data bus leads to an average energy consumption of up to 14.6%.

[1]  Massoud Pedram,et al.  ALBORZ: Address Level Bus Power Optimization , 2002, Proceedings International Symposium on Quality Electronic Design.

[2]  Cristina Silvano,et al.  Power optimization of system-level address buses based on software profiling , 2000, Proceedings of the Eighth International Workshop on Hardware/Software Codesign. CODES 2000 (IEEE Cat. No.00TH8518).

[3]  Massoud Pedram,et al.  Irredundant address bus encoding for low power , 2001, ISLPED '01.

[4]  Peter Petrov,et al.  Low-power instruction bus encoding for embedded processors , 2004, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[5]  Tomás Lang,et al.  Working-zone encoding for reducing the energy in microprocessor address buses , 1998, IEEE Trans. Very Large Scale Integr. Syst..

[6]  Behrooz Parhami,et al.  Computer arithmetic - algorithms and hardware designs , 1999 .

[7]  Kevin Skadron,et al.  Odd/even bus invert with two-phase transfer for buses with coupling , 2002, ISLPED '02.

[8]  Zhan Yu,et al.  The use of reduced two's-complement representation in low-power DSP design , 2002, 2002 IEEE International Symposium on Circuits and Systems. Proceedings (Cat. No.02CH37353).

[9]  Jan M. Rabaey,et al.  Architectural power analysis: The dual bit type method , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[10]  Alexander Albicki,et al.  Low power and high speed multiplication design through mixed number representations , 1995, Proceedings of ICCD '95 International Conference on Computer Design. VLSI in Computers and Processors.

[11]  Tughrul Arslan,et al.  A coefficient segmentation algorithm for low power implementation of FIR filters , 1999, ISCAS'99. Proceedings of the 1999 IEEE International Symposium on Circuits and Systems VLSI (Cat. No.99CH36349).

[12]  K. Banerjee,et al.  Scaling analysis of multilevel interconnect temperatures for high-performance ICs , 2005, IEEE Transactions on Electron Devices.

[13]  Nikil D. Dutt,et al.  Low power address encoding using self-organizing lists , 2001, ISLPED '01.

[14]  Mircea R. Stan,et al.  Bus-invert coding for low-power I/O , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[15]  Kiyoung Choi,et al.  Narrow bus encoding for low-power DSP systems , 2001, IEEE Trans. Very Large Scale Integr. Syst..

[16]  Anantha P. Chandrakasan,et al.  Bus Energy Reduction by Transition Pattern Coding Using a Detailed Deep Submicrometer Bus Model , 2003 .

[17]  Luca Benini,et al.  Asymptotic zero-transition activity encoding for address busses in low-power microprocessor-based systems , 1997, Proceedings Great Lakes Symposium on VLSI.