Approximate Multipliers Based on New Approximate Compressors

Esposito, Darjn; Strollo, A.G.M.; Napoli, Ettore; De, Davide; Petra, Nicola

doi:10.1109/tcsi.2018.2839266

Cited by 191 publications

(97 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Esposito et al [34] developed XOR-free inexact compressors that achieve a small error. The XOR-free compressors deliver a significant reduction in area utilization and power consumption.…”

Section: B Approximate Non-logarithmic Multipliersmentioning

confidence: 99%

On the Design of Logarithmic Multiplier Using Radix-4 Booth Encoding

Pilipović

Bulić

2020

IEEE Access

View full text Add to dashboard Cite

This paper proposes an energy-efficient approximate multiplier which combines radix-4 Booth encoding and logarithmic product approximation. Additionally, a datapath pruning technique is proposed and studied to reduce the hardware complexity of the multiplier. Various experiments were conducted to evaluate the multiplier's error performance and efficiency in terms of energy and area utilization. The reported results are based on simulations using TSMC-180nm. Also, the applicability of the proposed multiplier is examined in image sharpening and convolutional neural networks. The applicability assessment shows that the proposed multiplier can replace an exact multiplier and deliver up to a 75% reduction in energy consumption and up to a 50% reduction in area utilization. Comparative analysis with the state-of-the-art multipliers indicates the potential of the proposed approach as a novel design strategy for approximate multipliers. When compared to the state-of-the-art approximate non-logarithmic multipliers, the proposed multiplier offers smaller energy consumption with the same level of applicability in image processing and classification applications. On the other hand, some state-of-the-art approximate logarithmic multipliers exhibit lower energy consumption than the proposed multiplier but deliver significant performance degradation for the selected application cases. INDEX TERMS Approximate computing, arithmetic circuit design, booth encoding, logarithmic multipliers, multipliers, power-efficient processing, truncated multipliers.

show abstract

“…Esposito et al [34] developed XOR-free inexact compressors that achieve a small error. The XOR-free compressors deliver a significant reduction in area utilization and power consumption.…”

Section: B Approximate Non-logarithmic Multipliersmentioning

confidence: 99%

On the Design of Logarithmic Multiplier Using Radix-4 Booth Encoding

Pilipović

Bulić

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…After partial products are obtained, Wallace tree structure depicted in Figure 8 is built to sum all partial products. For the CSAs, we used an approximation methodology described in [38] to realize approximate (3:2) and (4:2) compressors. In final adder, designed approximate Sklansky adder are used.…”

Section: Approximate Multipliermentioning

confidence: 99%

Approximate CPU Design for IoT End-Devices with Learning Capabilities

2020

View full text Add to dashboard Cite

With the rise of Internet of Things (IoT), low-cost resource-constrained devices have to be more capable than traditional embedded systems, which operate on stringent power budgets. In order to add new capabilities such as learning, the power consumption planning has to be revised. Approximate computing is a promising paradigm for reducing power consumption at the expense of inaccuracy introduced to the computations. In this paper, we set forth approximate computing features of a processor that will exist in the next generation low-cost resource-constrained learning IoT devices. Based on these features, we design an approximate IoT processor which benefits from RISC-V ISA. Targeting machine learning applications such as classification and clustering, we have demonstrated that our processor reinforced with approximate operations can save power up to 23% for ASIC implementation while at least 90% top-1 accuracy is achieved on the trained models and test data set.

show abstract

“…The approximate 4-2 compressor [11] can be constructed by two XOR gates, one OR gate, and an AND-OR complex compound gate. In [12], the researchers analysed the error probability and the average error in a systematic way to design the approximate 4-2 compressor. The approximate 4-2 compressor [12] can be implemented by using two AND and two OR gates (without XOR gates) that can reduce the hardware complexity effectively.…”

Section: Imprecise 4-2 Compressorsmentioning

confidence: 99%

“…Approximate 4-2 compressor [12] has the least cost and the shortest critical path but it has the highest error rate. Note that the outputs of approximate 4-2 compressor [12] have the same weight. However, if the outputs of approximate 4-2 compressor [12] are different weights, the error of this compressor is 56.25%.…”

Section: Imprecise 4-2 Compressorsmentioning

confidence: 99%

“…However, some applications such as signal/image processing, communication, and multimedia can tolerate errors and produce results which are good enough for human perception [8][9][10]. Since approximate or lessthan-optimal results are sufficient, these error-tolerant applications allow computing systems to trade quality for energy, area, and performance [10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Imprecise 4‐2 compressor design used in image processing applications

Chang

Cheng

Lin

et al. 2019

IET Circuits, Devices & Systems

View full text Add to dashboard Cite

Approximate computing can be used in the error-tolerant applications since it can provide meaningful results with lower power consumption. In this study, the authors proposed a novel imprecise 4-2 compressor which is used in the multipliers of image processing applications. Besides the output values, they also consider the pattern distribution to resynthesise the 4-2 compressor in imprecise style. Compared to the precise compressor, the proposed imprecise 4-2 compressor can reduce power consumption and delay by 56 and 39%. Compared to the precise multiplier, the simulation results show that the multiplier which uses the proposed imprecise 4-2 compressor can achieve 33 and 30% improvement in power consumption and delay, respectively. In addition, the image quality of their design is good for human perception because peak signal-to-noise ratio PSNR values are more than 33 and mean structural similarity values are more than 0.99. Even compared to the related imprecise works, their design has a better error rate to improve the quality of images while maintains both the high power efficiency and low circuit complexity.

show abstract

Approximate Multipliers Based on New Approximate Compressors

Cited by 191 publications

References 29 publications

On the Design of Logarithmic Multiplier Using Radix-4 Booth Encoding

On the Design of Logarithmic Multiplier Using Radix-4 Booth Encoding

Approximate CPU Design for IoT End-Devices with Learning Capabilities

Imprecise 4‐2 compressor design used in image processing applications

Contact Info

Product

Resources

About