Error Analysis of Certain Floating-Point On-Line Algorithms
- 1 April 1983
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. C-32 (4) , 352-358
- https://doi.org/10.1109/tc.1983.1676236
Abstract
The properties of redundant number system in significand (mantissa) representation are studied and the range of redundant significand is derived. From the range of the redundant significand and the absolute error of on-line operations, the MRRE (maximum relative representation error) is defined and analyzed for floating-point on-line addition and multiplication.Keywords
This publication has 15 references indexed in Scilit:
- An On-Line Square Root AlgorithmIEEE Transactions on Computers, 1982
- Floating-point on-line arithmetic: AlgorithmsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1981
- Design of a digit-slice on-line arithmetic unitPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1981
- Significance Arithmetic: Application to a Partial Differential EquationIEEE Transactions on Computers, 1977
- On-Line Algorithms for Division and MultiplicationIEEE Transactions on Computers, 1977
- On the Precision Attainable with Various Floating-Point Number SystemsIEEE Transactions on Computers, 1973
- A Mean Square Estimate of the Generated Roundoff Error in Constant Matrix Iterative ProcessesJournal of the ACM, 1971
- On the Distribution of NumbersBell System Technical Journal, 1970
- Representation Error for Real Numbers in Binary Computer ArithmeticIEEE Transactions on Electronic Computers, 1967
- Significant Digit Computer ArithmeticIEEE Transactions on Electronic Computers, 1958