PROJECT TITLE :

Energy-Efficient VLSI Realization of Binary64 Division With Redundant Number Systems - 2017

ABSTRACT:

VLSI realizations of digit-recurrence binary division typically use redundant illustration of partial remainders and quotient digits. The former permits for quick carry-free computation of the next partial remainder, and also the latter ends up in less number of the required divisor multiples. In finding out the previous relevant works, we have a tendency to have noted that the binary carry-save (CS) variety system is prevalent within the representation of partial remainders, and redundant high radix representation of quotient digits is widespread so as to reduce the cycle count. During this paper, we have a tendency to explore a style house containing four division architectures. These are based on binary CS or radix-16 signed digit (SD) representations of partial remainders. On the other hand, they use full or partial precomputation of divisor multiples. The latter uses smaller multiplexer at the price two extra adders, where one among the operands is constant at intervals all cycles. The quotient digits are represented by radix-sixteen [-9, nine] SDs. Our synthesis-based analysis of VLSI realizations of the most effective previous relevant work and the four proposed styles show reduced power and energy figures in the proposed designs at the cost of more silicon area and delay measures. But, our energy-delay product is 26%-thirty fivepercent but that of the reference work.