DocumentCode :
1307804
Title :
Residue-to-binary converters based on new Chinese remainder theorems
Author :
Wang, Yuke
Author_Institution :
Dept. of Comput. Sci. & Eng., Florida Atlantic Univ., Boca Raton, FL, USA
Volume :
47
Issue :
3
fYear :
2000
fDate :
3/1/2000 12:00:00 AM
Firstpage :
197
Lastpage :
205
Abstract :
The speed of arithmetic operations depends on the size of the numbers involved. Smaller numbers have faster operations. That is exactly the reason why residue number systems are attractive in computer arithmetic. However, the conversion from residue to binary numbers involves a large number module operations. Several residue-to-binary converters are proposed in this paper. The converters are based on the New Chinese Remainder Theorems (CRT´s) I and II which represent our work. The New CRT´s improve the celebrated CRT. The new algorithms do not use any large size module adders. The numbers involved are much smaller compared to the numbers in the CRT and its alternative, the Mixed Radix Conversion method. Given a moduli set as (P1, P 2,, Pn), to convert a residue number (x1,x2,...,xn) to its decimal correspondence, a matrix of numbers bounded by Pi is needed for the New CRT I compared to the large numbers M/Pi for the CRT, where M=P1P2...Pn. The New CRT II uses module multipliers of size less than √M. If the condition P i+1>P1+P2+...+Pi is satisfied, only one module operation of size Pn is needed for the conversion. Residue-to-binary conversion based on the New CRT´s presented here will have a significant impact on many algorithms which currently use the CRT, particularly in computer arithmetic such as residue number systems
Keywords :
data conversion; residue number systems; Chinese remainder theorems; arithmetic operations; computer arithmetic; module multipliers; module operation; residue number systems; residue-to-binary converters; Adders; Cathode ray tubes; Circuits and systems; Digital arithmetic; Digital signal processing; Hardware; Image converters; Limiting; Matrix converters; Signal processing algorithms;
fLanguage :
English
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1057-7130
Type :
jour
DOI :
10.1109/82.826745
Filename :
826745
Link To Document :
بازگشت