An architecture for parallel multipliers

A novel architecture for parallel multipliers is proposed based on a previously published architecture that used five-counters as the basic building block. The architecture removes the redundancy associated with an incompletely spanned output range of the five-counter cell, by changing the building block to a two-bit full adder. This cell has the same number of inputs, but produces an output that completely spans the output dynamic range. It is shown that a dynamic logic implementation of the five-counter has a 46% hardware increase over the two-bit full adder cell proposed for this architecture. The resulting two-bit adder cell architecture shows an almost 50% decrease in silicon area and almost a 30% decrease in evaluation time compared to the five-counter cell architecture.<<ETX>>

[1]  Shinji Nakamura Algorithms for Iterative Array Multiplication , 1986, IEEE Transactions on Computers.

[2]  Shinji Nakamura,et al.  A Single Chip Parallel Multiplier by MOS Technology , 1988, IEEE Trans. Computers.

[3]  Paul A. Wintz,et al.  Fast Multipliers , 1970, IEEE Transactions on Computers.

[4]  Earl E. Swartzlander,et al.  Arithmetic for digital neural networks , 1991, [1991] Proceedings 10th IEEE Symposium on Computer Arithmetic.

[5]  C. W. Slayman,et al.  A high-speed high-density silicon 8/spl times/8-bit parallel multiplier , 1987 .

[6]  Christopher S. Wallace,et al.  A Suggestion for a Fast Multiplier , 1964, IEEE Trans. Electron. Comput..

[7]  Christer Svensson,et al.  High-speed CMOS circuit technique , 1989 .

[8]  David L. Pulfrey,et al.  Design procedures for differential cascode voltage switch circuits , 1986 .

[9]  B. Maden,et al.  Parallel architectures for high speed multipliers , 1989, IEEE International Symposium on Circuits and Systems,.