Process variation-aware approximate full adders for imprecision-tolerant applications. (October 2020)

• Record Type: Journal Article
• Authors: Mirzaei, Mohammad; Mohammadi, Siamak
• Is Part Of: Computers & electrical engineering. Volume 87(2020)
• Publication Date: 2020-10
• Subjects: Imprecision-tolerant applications -- Approximate computing -- Full adder -- Process variation; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2020.106761
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: We show pass transistorbased full adders are the most sensitive to variations. All CMOS approximate full adders have shown less variability. The adders error percentage in the presence of variation is considered. Proposed adder presents the best tradeoff in terms of power, delay, area and PSNR. Graphical abstract: Abstract: In imprecision-tolerant applications such as multimedia and signal processing a slightly degraded output quality is acceptable, which could lead to significant power reduction. We have proposed three approximate full adders for such applications. We have also considered the process variation impacts on approximate full adders for the first time. Twelve approximate full adders including ours have been used and analyzed in a ripple carry adder, as well as in a Sobel edge detection application. Experimental results show our design compared to the best existing approximate full adder reduces power consumption, delay and power-delay-product by 6.2%, 42.3% and 46.02%, respectively. Our proposed approximate ripple carry adder has the best performance in terms of error parameters, and the best PSNR (Peak signal-to-noise ratio) when used in Sobel algorithm. When the proposed approximate full adder is used in Sobel algorithm, it exhibits the least variability in terms of delay compared to other counterparts.