Paper Title Low Cost and Low Power Floating-point Fused Multiply-Add Unit Design with Proxy Bits and Weighted 2-Level Booth Encoding
نویسندگان
چکیده
With the appearance of high performance mobile devices, low cost and low power consumption have become important issues in high performance processors. To meet the needs, low cost and low power floating-point fused multiply-add unit is proposed in this paper. According to the area and power consumption analysis, the multiplication part in fused multiplyadd operation accounted for most power consumption and area. Thus, the proposed floating-point fused multiply-add unit used a new weighted 2-level Booth encoding algorithm to optimize the multiplier. In additional, proxy bits, which can represent redundant bits, from the cancellation in subtraction operation is proposed. The aim of this paper was to be achieved by optimizing a shifter, leading zero anticipator, and adder using the proxy bits. The synthesis and power consumption analysis results show that the proposed floating-point fused multiply-add unit reduced area by 37.3%, and improves the latency by 10.8% compared to a conventional floating-point fused multiply-add unit under no clock constraints. In addition, the proposed floating-point fused multiply-add unit reduced area and power consumption compared to the conventional unit by 31.6% and 23.3%, respectively under 2.5ns timing constraints. Therefore, the proposed floating-point fused multiply-add unit will contribute to server and mobile purpose processors’ high performance, low power, and low cost requirements.. (Abstract) KeywordsFloating-point unit, Fused multiply-add unit, Low power, Low cost, Weighted 2-level Booth encoding, Proxy bits
منابع مشابه
ضربکننده و ضربجمعکننده پیمانه 2n+1 برای پردازنده سیگنال دیجیتال
Nowadays, digital signal processors (DSPs) are appropriate choices for real-time image and video processing in embedded multimedia applications not only due to their superior signal processing performance, but also of the high levels of integration and very low-power consumption. Filtering which consists of multiple addition and multiplication operations, is one of the most fundamental operatio...
متن کاملLow Power Floating-Point Multiplier Based On Vedic Mathematics
Fast Fourier transform (FFT) coprocessor, having significant impact on the performance of communication systems, hasbeen a hot topic of research for many years. The FFT function consists of consecutive multiply add operations over complex numbers, dubbed as butterfly units. Applying floating-point (FP) arithmetic to FFT architectures, specifically butterfly units, has become more popular recent...
متن کاملFloating-Point Single-Precision Fused Multiplier-adder Unit on FPGA
The fused multiply-add operation improves many calculations and therefore is already available in some generalpurpose processors, like the Itanium. The optimization of units dedicated to execute the multiply-add operation is therefore crucial to achieve optimal performance when running the overlying applications. In this paper, we present a single-precision floating-point fused multiply-add opt...
متن کاملDesign and Simulation of a 2GHz, 64×64 bit Arithmetic Logic Unit in 130nm CMOS Technology
The purpose of this paper is to design a 64×64 bit low power, low delay and high speed Arithmetic Logic Unit (ALU). Arithmetic Logic Unit performs arithmetic operation like addition, multiplication. Adders play important role in ALU. For designing adder, the combination of carry lookahead adder and carry select adder, also add-one circuit have been used to achieve high speed and low area. In mu...
متن کاملModified Multiply and Accumulate Unit with Hybrid Encoded Reduced Transition Activity Technique Equipped Multiplier and Low Power 0.13μm Adder for Image Processing Applications
This paper explores the design approach of a low power high performance Multiply and Accumulate (MAC) unit with hybrid encoded Reduced Transition Activity Technique (RTAT) equipped multiplier and low power 0.13μm adder. Design of a low power MAC unit for image processing systems exploiting insignificant bits in pixels values and the similarity of neighboring pixels in video streams is presented...
متن کامل