A Radix-16 Booth Multiplier Based on Recoding Adder with Ultra High Power Efficiency and Reduced Complexity for Neuroimaging

Abstract

In the current medical developments the neuro imaging plays a vital role in the study of a human brain related disorders. The accuracy of the brain study is mainly dependent on the images created from the scanners at a rapid speed. In achieving this we need a high speed and low power consuming scanners. The current scenario in VLSI design, the scanners highly rely on a high speed Digital Signal Processor (DSP), which generally depends on the speed of a multiplier. Multipliers are considered as a more complex component when compared with adders. The current techniques provide greater access to high-speed multipliers which are designed with less area that consume low power. The major constraints to be considered for an efficient multiplier design are propagation time delay and power dissipation, especially during the ideal time. An approximate recoding adder is proposed to reduce the existing booth multiplier's immensity. It increases the accuracy and reduces complexity through this technique; however, it has an issue with Power Delay Product (PDP) and power dissipation. To solve this problem, the proposed system is designed with a power gating based 16 × 16 bit Booth multiplier based on approximate recoding adder. It decreases the power dissipation and minimizes the length and width of the partial products for speeding up the multiplication process. The results obtained from the simulation show that the designed power gating based Radix multiplier circuits achieves better PDP, average power and area. The achieved results are compared with a Radix based multiplier, power gating CLA based multiplier and CLA based multiplier.