Subscribe

Issue No.11 - Nov. (2012 vol.61)

pp: 1551-1562

Álvaro Vázquez , Dept. of Electron. & Comput. Sci., Univ. of Santiago, Santiago, Spain

J. Villalba-Moreno , Dept. of Comput. Archit., Univ. of Malaga, Malaga, Spain

E. Antelo , Dept. of Electron. & Comput. Sci., Univ. of Santiago, Santiago, Spain

E. L. Zapata , Dept. of Comput. Archit., Univ. of Malaga, Malaga, Spain

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TC.2011.217

ABSTRACT

In this work, we propose a new decimal redundant CORDIC algorithm to manage transcendental functions, using floating-point representation. The algorithms determine the direction of the elementary rotation using sign estimations. Unlike binary redundant CORDIC, repetition of iterations are not required to ensure convergence since novel decimal codes have been carefully selected with sufficient redundancy to prevent any repetition. The algorithms are mapped to a low-cost unit based on a decimal 3-2 carry-save adder which can also be used as a floating-point decimal division unit. Compared to current decimal floating-point units, the implementation of our algorithm involves minor modifications of the native hardware, while providing a huge set of elementary functions.

INDEX TERMS

floating point arithmetic, decimal floating-point units, redundant floating point decimal CORDIC algorithm, transcendental functions, floating-point representation, decimal codes, Estimation, Convergence, Adders, Redundancy, Vectors, Approximation algorithms, Hardware, CORDIC algorithm, Estimation, Convergence, Adders, Redundancy, Vectors, Approximation algorithms, Hardware, decimal arithmetic, Estimation, Convergence, Adders, Redundancy, Vectors, Approximation algorithms, Hardware, carry-save arithmetic

CITATION

Álvaro Vázquez, J. Villalba-Moreno, E. Antelo, E. L. Zapata, "Redundant Floating-Point Decimal CORDIC Algorithm",

*IEEE Transactions on Computers*, vol.61, no. 11, pp. 1551-1562, Nov. 2012, doi:10.1109/TC.2011.217REFERENCES

- [1] IEEE,
IEEE Standard for Floating-Point Arithmetic, IEEE Std 754-2008, Aug. 2008.- [2] L. Eisen et al., “IBM Power 6 Accelerators: VMX and DFU,”
IBM J. Research and Development, vol. 51, no. 6, pp. 663-684, Nov. 2007.- [3] E.M. Schwarz and S.R. Carlough, “Power 6 Decimal Divide,”
Proc. IEEE Int'l Conf. Application-Specific Systems, Architectures and Processors, pp. 128-133, 2007.- [4] E.M. Schwarz et al., “Decimal Floating-Point Support on the IBM System Z10 Processor,”
IBM J. Research and Development, vol. 53, no. 1, pp. 4:1-4:10, 2009.- [5] M. Cornea, J. Harrison, C. Anderson, P.T.P. Tang, E. Schneider, and E. Gvozdev, “A Software Implementation of the IEEE 754R Decimal Floating-Point Arithmetic Using the Binary Encoding Format,”
IEEE Trans. Computers, vol. 58, no. 2, pp. 148-162, Feb. 2009.- [6] J. Harrison, “Decimal Transcendentals via Binary,”
Proc. IEEE 19th Symp. Computer Arithmetic, pp. 187-194, 2009.- [7] A. Vázquez and E. Antelo, “Constant Factor CORDIC Algorithm for Decimal BCD Input Operands,”
Proc. Eighth Real Numbers and Computers Conf., July 2008.- [8] A. Vázquez, J. Villalba, and E. Antelo, “Computation of Decimal Transcendental Functions Using the CORDIC Algorithm,”
Proc. IEEE 19th Symp. Computer Arithmetic, pp. 179-186, 2009.- [9] M. Ercegovac and T. Lang,
Digital Arithmetic. Morgan Kaufmann, 2004.- [10] R.K. Richards,
Arithmetic Operations in Digital Computers. Van Nostrand, 1955.- [11] S. Gorgin and G. Jaberipur, “Fully Redundant Decimal Arithmetic,”
Proc. IEEE 19th Symp. Computer Arithmetic pp. 145-152, 2009.- [12] J.M. Muller,
Elementary Functions: Algorithms and Implementation, second ed. Birkhauser Verlag, 2007.- [13] N. Takagi and T. Asada, and S. Yajima, “Redundant CORDIC Methods with a Constant Scale Factor,”
IEEE Trans. Computers, vol. 40, no. 9, pp. 989-995, Sept. 1991.- [14] J.A. Lee and T. Lang, “Constant-Factor Redundant CORDIC for Angle Calculation and Rotation,”
IEEE Trans. Computers, vol. 41, no. 8, pp. 1016-1025, Aug. 1992.- [15] A. Vázquez, J. Villalba, E. Antelo, and E.L. Zapata, “Calculations for Redundant Floating-point Decimal CORDIC Algorithm,” Internal Report, Univ. of Santiago de Compostela (www. ac.usc.es/system/filesadditional_material_CORDIC4.pdf) , 2011.
- [16] J.S. Walther,
Elementary Floating Point CORDIC Function Processor and Shifter, US Patent 3766370, Washington, D.C.: Patent and Trademark Office, May 14, 1973.- [17] H.M. Ahmed, “Efficient Elementary Function Generation with Multipliers,”
Proc. Ninth Symp. Computer Arithmetic, pp. 52-59, 1989.- [18] J.E. Meggitt, “Pseudo Division and Pseudo Multiplication Processes,”
IBM J. Research and Development, vol. 6, pp. 210-226, Apr. 1962.- [19] D.S. Cochran, “Algorithms and Accuracy in the HP-35,”
Hewlett-Packard J., vol. 23, pp. 10-11, 1972.- [20] H. Schmid,
Decimal Computation. John Wiley & Sons, 1974.- [21] L. Imbert, J.M. Muller, and F. Rico, “A Radix-10 BKM Algorithm for Computing Transcendentals on Pocket Computers,”
J. VLSI Signal Processing Systems, vol. 25, no. 2, pp. 179-186, June 2000.- [22] A. Jimeno et al., “A BCD-Based Architecture for Fast Coordinate Rotation,”
J. Systems Architecture, vol. 53, no. 8, pp. 829-840, 2008. |