An architecture for high speed Radix10 division

Decimal arithmetic is gaining more and more importance in business, commercial and financial applications due to error free and high speed computations. In this work, high speed radix10 divider architecture has been proposed to reduce the delay. This paper presents a modified architecture in which intermediate results are utilized to perform the high speed division. The modified architecture is simulated for different numbers of bits. Synthesis results show that the modified architecture implemented in 180nm technology has reduced delay when compared to digit recurrence with constant digit selection function architecture which is the fastest of existing architectures.

[1]  Michael J. Schulte,et al.  A Decimal Floating-Point Divider Using Newton–Raphson Iteration , 2007, J. VLSI Signal Process..

[2]  Paolo Montuschi,et al.  A radix-10 SRT divider based on alternative BCD codings , 2007, 2007 25th International Conference on Computer Design.

[3]  Michael J. Flynn,et al.  Division Algorithms and Implementations , 1997, IEEE Trans. Computers.

[4]  Milos D. Ercegovac,et al.  Design and FPGA implementation of radix-10 combined division/square root algorithm with limited precision primitives , 2010, 2010 Conference Record of the Forty Fourth Asilomar Conference on Signals, Systems and Computers.

[5]  Standard for Floating-Point Arithmetic , 2018 .

[6]  Tomás Lang,et al.  Digit-recurrence dividers with reduced logical depth , 2005, IEEE Transactions on Computers.

[7]  Malte Baesler,et al.  A radix-10 digit recurrence division unit with a constant digit selection function , 2010, 2010 IEEE International Conference on Computer Design.

[8]  Tomás Lang,et al.  A Radix-10 Digit-Recurrence Division Unit: Algorithm and Architecture , 2007, IEEE Transactions on Computers.