检索结果-内蒙古大学图书馆

IEEE MICRO 2018年第4期38卷 8-10页

作者： Jerger, Natalie Enright San Miguel, Joshua Univ Toronto Elect & Comp Engn Toronto ON Canada Univ Wisconsin Elect & Comp Engn Madison WI 53706 USA

This special issue of IEEE Micro explores exciting, new ideas in the vast design space of approximate computing. We present articles that range from programming languages to circuits and cover important application do... 详细信息

关键词： Special issues and sections Internet of Things Machine learning approximate computing Computer Programming Languages approximate computing Machine Learning Internet of things Artificial Intelligence

来源：评论

学校读者我要写书评

暂无评论

approximate computing

引用

IT-INFORMATION TECHNOLOGY 2022年第3期64卷 77-78页

作者： Keszocze, Oliver Friedrich Alexander Univ Erlangen Nurnberg FAU Dept Comp Sci D-91058 Erlangen Germany

来源：评论

学校读者我要写书评

暂无评论

Improve accuracy in CNNs while using approximate computing methods

引用

JOURNAL OF SUPERcomputing 2025年第2期81卷 1-23页

作者： Rafieinejad, Mohammadreza Marvasti, Mohammadreza Binesh Asghari, Seyyed Amir Shahbakhti, Kimiya Kharazmi Univ Dept Elect & Comp Engn Tehran Iran

Artificial neural networks have been one of the science's most influential and essential branches in the past decades. Neural networks have found applications in various fields including medical and pharmaceutical services, voice and speech recognition, computer vision, natural language processing, and video and image processing. Neural networks have many layers and consume much energy. approximate computing is a promising way to reduce energy consumption in applications that can tolerate a degree of accuracy reduction. This paper proposes an effective method to prevent accuracy reduction after using approximate computing methods in the CNNs. The method exploits the k-means clustering algorithm to label pixels in the first convolutional layer. Then, using one of the existing pruning methods, different pruning amounts have been applied to all layers. The experimental results on three CNNs and four different datasets show that the accuracy of the proposed method has significantly improved (by 17%) compared to the baseline network.

关键词： approximate computing Computer vision Convolutional neural network Energy consumption Machine learning Neural network

来源：评论

学校读者我要写书评

暂无评论

Implementation of Energy-Efficient approximate computing via Recursive Multiplier in Error-Tolerant Applications

引用

JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS 2025年第7期34卷

作者： Karthikkumar, M. Kalaivani, R. Erode Sengunthar Engn Coll Dept Elect & Commun Engn Thuduppathi 638057 Tamil Nadu India

approximate computing (AC) in arithmetic logic has become a viable option in applications requiring error tolerance in energy-efficient architectures. AC relied on approximate arithmetic functions to reduce delay, area, and power consumption while sacrificing accuracy to reduce delay, power, and area. In this research, a novel two approximate recursive multipliers (RMul-1, RMul-2) and an approximate adder have been designed to reduce power consumption, area, and computational delay in error-tolerant systems. The recursive multipliers utilize a combination of NOR, AND, half adder, and full adder gates to achieve low power and area-efficient designs. Furthermore, to reduce time, the approximation adder uses an optimal combination of AND, OR, and MUX gates. Furthermore, the cadence RTL compiler synthesizes the proposed multiplier using 28nm technology, and it is compared to previous approximation multipliers. Image processing applications are simulated, and the performance of the proposed multipliers is verified using simulations using the Xilinx ISE 13.2 tool. The proposed RMul designs outperform current techniques by up to 30.3% in area, 20.2% in power, and 43.9% in delay, according to experimental results. In addition, the suggested multipliers outperform existing multipliers in terms of SSIM and PSNR.

关键词： approximate computing approximate adder approximate recursive multiplier power consumption

来源：评论

学校读者我要写书评

暂无评论

Self-aware decentralized security for real time approximate computing tasks in FPGA-based edge platforms

引用

JOURNAL OF SUPERcomputing 2025年第1期81卷 1-45页

作者： Guha, Krishnendu Univ Coll Cork Sch Comp Sci & Informat Technol Cork Ireland

The present era has witnessed the wide deployment of reconfigurable hardware or Field Programmable Gate Arrays (FPGAs) in edge and cloud platforms. With its ability of dynamic partial reconfiguration at runtime, FPGAs provide the apt environment to execute a variety of real-time tasks in strict power and timing constraints. However, threats associated with the vulnerability of hardware like hardware Trojan horses may cause sudden delays at runtime or may even drain the power budget of the system to prevent completion of the tasks before their associated deadlines. We consider a resource-constraint FPGA-based edge platform with strict power budget. This is associated with execution of several periodic and nonperiodic hard real-time approximate computing tasks, i.e., tasks whose result can vary within a certain range but must complete within a prespecified deadline. We depict how delay inducing and power draining hardware Trojans may jeopardize the scenario. We propose deployment of low overhead agents or self-aware modules (SAMs) that can facilitate decentralized control and nonintrusive security in such an environment. With each FPGA that is entrusted with execution of a series of tasks or a task schedule, a SAM is associated. The SAM continuously monitors the performance of its host, based on prespecified power and timing data. On detecting any anomaly, it outsources the tasks to other SAMs for execution in other FPGAs, so that the tasks can complete their execution prior to their deadline. Low resource utilization and timing overhead of SAM, high task success rate for periodic tasks and low task rejection rate for nonperiodic tasks depict the suitability of our proposed mechanism.

关键词： Security Real-time tasks approximate computing Hardware Trojans

来源：评论

学校读者我要写书评

暂无评论

Multiplierless QRS detection algorithm for mobile ECG monitoring based on approximate computing

引用

MEASUREMENT 2025年 249卷

作者： Reklewski, Wojciech Heryan, Katarzyna Augustyniak, Piotr Miskowicz, Marek AGH Univ Krakow Dept Biocybernet & Biomed Engn Krakow Poland AGH Univ Krakow Dept Measurement & Elect Krakow Poland AGH Univ Krakow Al Mickiewicza 30 PL-30059 Krakow Poland

The QRS detectors for wearable devices should be energy-efficient, even at the price of some loss of detection accuracy. This paper presents a low-complexity multiplierless QRS complex detection algorithm for mobile longterm ECG monitoring based on arithmetic that approximates real numbers by a sum of power-of-two components. The proposed hardware-friendly computation technique involves only additions, subtractions, and multiplications or divisions by powers-of-two, executed simply by bit-shift operations. The performance of the algorithm, called Multiplierless Dual-Path Preprocessing (ML-DPP) QRS detector, on the MIT-BIH Database is as follows: Se = 99.66 %, PPV = 99.52 %, F1 = 99.59 %, ACC = 99.18 %, and DER = 0.82 %. The deterioration in detection accuracy of the multiplierless version is negligible compared to the optimal algorithm settings with real numbers, while computational complexity is substantially reduced. The proposed methodology to support the multiplication-free arithmetic can be applied to other QRS detection algorithms.

关键词： QRS detection Wearable ECG devices approximate computing Sum of power-of-two components Multiplication-free arithmetic Multiplierless FIR filters

来源：评论

学校读者我要写书评

暂无评论

approximate computing

引用

IEEE NANOTECHNOLOGY MAGAZINE 2022年第1期16卷 10-10页

作者： Han, Jie Liu, Weiqiang

Nanotechnology has been plagued by reliability issues when used in building circuits and systems for computing. approximate computing, on the other hand, exploits the inherent tolerance of inaccuracies and imperfections in many applications for performance and hardware efficiency. Will approximate computing provide a remedy for emerging nanotechnologies? Five research groups around the globe have made contributions in this special issue to address some key issues in building approximate computing systems using nanotechnologies.

关键词： Special issues and sections Reliability approximate computing Performance evaluation

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient approximate computing Framework for DNN Acceleration Using a Probabilistic-Oriented Method

引用

IEEE Transactions on Emerging Topics in computing 2025年

作者： Huang, Pengfei Chen, Ke Wang, Chenghua Liu, Weiqiang Nanjing University of Aeronautics and Astronautics College of Electronic and Information Engineering College of Integrated Circuits Nanjing211106 China

approximate computing (AxC) has recently emerged as a successful approach for optimizing energy consumption in error-tolerant applications, such as deep neural networks (DNNs). The enormous model size and high computation cost of DNNs present significant challenges for deployment in energy-efficient and resource-constrained computing systems. Emerging DNN hardware accelerators based on AxC designs selectively approximate the non-critical segments of computation to address these challenges. However, a systematic and principled approach that incorporates domain knowledge and approximate hardware for optimal approximation is still lacking. In this paper, we propose a probabilistic-oriented AxC (PAxC) framework that provides high energy savings with acceptable quality by considering the overall probability effect of approximation. To achieve aggressive approximate designs, we utilize the minimum likelihood error to determine the AxC synergy profile at both application and circuit levels. This enables effective coordination of the trade-off between energy and accuracy. Compared with a baseline design, the power-delay product (PDP) is significantly reduced by up to 83.66% with an acceptable accuracy reduction. Simulation and a case study of the image process validate the effectiveness of the proposed framework. © 2013 IEEE.

关键词： approximate computing DNN accelerator hardware acceleration optimized approximate circuits probabilistic AxC

来源：评论

学校读者我要写书评

暂无评论

Innovative Approaches to Low-Cost Multiplier Design Using approximate computing and Clock Gating

Innovative Approaches to Low-Cost Multiplier Design Using Ap...

引用

Device Intelligence, computing and Communication Technologies, (DICCT), International Conference on

作者： Nihar Buxy Shilpa D R Dept. of Electronics and Communication Eng. R.V. College of Engineering Bengaluru India

ISBN: (数字)9798331543358

ISBN: (纸本)9798331543365

In this paper, we propose a novel design methodology for low-cost approximate radix-4 booth multipliers that significantly reduce energy consumption in error-tolerant signal processing applications. Unlike prior works that focus solely on either partial product generation (PPG) or partial product accumulation (PPA), our approach co-designs these units to balance approximation errors, resulting in an internal error mean that is nearly zero. This integration enables substantial energy savings while maintaining comparable accuracy to existing designs. To further enhance energy efficiency, we incorporate a Synthesized-based clock gating technique, reducing unnecessary switching activity in the design. Experimental evaluations on Finite Impulse Response (FIR) filtering and image classification tasks demonstrate that the proposed multiplier achieves up to 39.3% reduction in total power consumption while maintaining accuracy within an acceptable range for error-tolerant applications. Additionally, delay and area utilisation are reduced by approximately 15.18% and 8.26%, respectively, compared to the existing method. Compared to traditional accurate booth multipliers, our design provides a better trade-off between power efficiency and computational precision, making it a promising alternative for energy-constrained digital signal processing (DSP) systems.

关键词： Performance evaluation Accuracy Power demand approximate computing Digital signal processing Switches Energy efficiency Delays Clocks Image classification

来源：评论

学校读者我要写书评

暂无评论

UNApprox: An Open-Source Tool for Loop Perforation in approximate computing

UNApprox: An Open-Source Tool for Loop Perforation in Approx...

引用

Latin American Test Workshop, LATW

作者： Alexander Aponte-Moreno Felipe Restrepo-Calle Cesar Pedraza Bonilla Dept. of Mechatronics Engineering Universidad Militar Nueva Granada Bogotá Colombia Dept. of Systems and Industrial Eng. Universidad Nacional de Colombia Bogotá Colombia

ISBN: (数字)9781665477635

ISBN: (纸本)9781665477642

Loop perforation is a well-known software-based approximate computing technique that improves efficiency at the expense of accuracy. In this paper, we present UNApprox, an open-source software tool that facilitates the easy application of loop perforation within an application's source code. As an open-source tool, UNApprox can also be extended to implement other approximate computing techniques. We tested the tool on three iterative algorithms, and the results are discussed in this paper.

关键词： Accuracy Program processors Source coding approximate computing Computer architecture Iterative methods Software tools Open source software

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：