Design of a low-power, high-speed 64-bit base-4 modified Booth multiplier in FPGA
Design of a low-power, high-speed 64-bit base-4 modified Booth multiplier in FPGA
Sajjad Babaei1
1) Master s degree student in Electrical and Electronics at Mohaghegh Ardabili University
IOI Code
:
AAHD-ABKZKA
Publication :
National Conference on Information Technology, Nanotechnology, Artificial Intelligence and Technological Futures Studies(itnaf.ir)
Abstract :
Objective: Multiplication operations are present in many parts of a digital system or digital computer. Especially in signal processing, graphics and scientific computing. With the advancements in technology, various techniques have been proposed for designing multipliers that have features such as speed, low power consumption and less area. Therefore, speed, compact VLSI implementation and low power consumption. These three parameters always reduce the power, area and speed of the values.
Research Methodology: This paper is dedicated to designing a high-speed and low-power Booth multiplier and implementing it on reconfigurable hardware for arithmetic multiplication, Booth multiplier architectures are discussed and found to be the most efficient among all and have optimal latency, low power for low power multiplication, Booth encoder switching and compressor placement techniques are used to reduce power and delay.
Findings: In digital systems, multiplication operations are very important, especially in applications such as image processing. This thesis proposes an approximate radix-4 Booth multiplier designed to increase performance and reduce complexity compared to similar examples, specifically for signed and unsigned numbers, the proposed multiplier is covered. Evaluation is done through simulation using Xilinx software with FPGA board. Using techniques such as modified Booth coding and parallel partial product generation, the aim is to achieve reduced latency and power consumption. The application of this multiplier is demonstrated. Quantitative criteria, including multiplier speedup and latency reduction, are presented. In addition, VHDL coding of Booth multipliers for 64 bits and their FPGA implementation is done by Xilinx project Navigator. The output is displayed. These synthesis results show that the computation time for the 64-bit product is 87.8 ns, and compared to other multipliers, the delay and power consumption reduction is evident.
Keywords : objective multiplication operations are present in many parts of a digital system or digital computer especially in signal processing graphics and scientific computing with the advancements in technology various techniques have been proposed for designing multipliers that have features such as speed low power consumption and less area therefore speed compact vlsi implementation and low power consumption these three parameters always reduce the power area and speed of the values research methodology this paper is dedicated to designing a high speed and low power booth multiplier and implementing it on reconfigurable hardware for arithmetic multiplication booth multiplier architectures are discussed and found to be the most efficient among all and have optimal latency low power for low power multiplication booth encoder switching and compressor placement techniques are used to reduce power and delay findings in digital systems multiplication operations are very important especially in applications such as image processing this thesis proposes an approximate radix 4 booth multiplier designed to increase performance and reduce complexity compared to similar examples specifically for signed and unsigned numbers the proposed multiplier is covered evaluation is done through simulation using xilinx software with fpga board using techniques such as modified booth coding and parallel partial product generation the aim is to achieve reduced latency and power consumption the application of this multiplier is demonstrated quantitative criteria including multiplier speedup and latency reduction are presented in addition vhdl coding of booth multipliers for 64 bits and their fpga implementation is done by xilinx project navigator the output is displayed these synthesis results show that the computation time for the 64 bit product is 87 8 ns and compared to other multipliers the delay and power consumption reduction is evident