Skip to Content

42 Results Found

  • Article
  • Open Access
1 Citations
4,826 Views
16 Pages

A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator

  • Yuhang Li,
  • Mei Wen,
  • Jiawei Fei,
  • Junzhong Shen and
  • Yasong Cao

15 September 2022

The systolic array provides extremely high efficiency for running matrix multiplication and is one of the mainstream architectures of today’s deep learning accelerators. In order to develop efficient accelerators, people usually employ simulato...

(This article belongs to the Topic Artificial Intelligence Models, Tools and Applications)
  • Article
  • Open Access
15 Citations
9,596 Views
13 Pages

High-Frequency Systolic Array-Based Transformer Accelerator on Field Programmable Gate Arrays

  • Yonghao Chen,
  • Tianrui Li,
  • Xiaojie Chen,
  • Zhigang Cai and
  • Tao Su

The systolic array is frequently used in accelerators for neural networks, including Transformer models that have recently achieved remarkable progress in natural language processing (NLP) and machine translation. Due to the constraints of FPGA EDA (...

(This article belongs to the Section Artificial Intelligence Circuits and Systems (AICAS))
  • Article
  • Open Access
2,195 Views
21 Pages

FPGA-based convolutional accelerators have been widely used in image recognition scenarios. Many convolutional accelerators utilize the systolic array structure to enhance parallelism. Developing a method to efficiently estimate the utilized hardware...

(This article belongs to the Special Issue FPGA-Based Reconfigurable Embedded Systems)
  • Article
  • Open Access
2 Citations
1,967 Views
16 Pages

Architecture Design of a Convolutional Neural Network Accelerator for Heterogeneous Computing Based on a Fused Systolic Array

  • Yang Zong,
  • Zhenhao Ma,
  • Jian Ren,
  • Yu Cao and ,
  • Meng Li and
  • Bin Liu
  • + 1 author+ 0 author

16 January 2026

Convolutional Neural Networks (CNNs) generally suffer from excessive computational overhead, high resource consumption, and complex network structures, which severely restrict the deployment on microprocessor chips. Existing related accelerators only...

(This article belongs to the Section Intelligent Sensors)
  • Article
  • Open Access
5 Citations
5,168 Views
15 Pages

27 November 2024

Transformer model is a type of deep learning model that has quickly become fundamental in natural language processing (NLP) and other machine learning tasks. Transformer hardware accelerators are usually designed for specific models, such as Bidirect...

(This article belongs to the Topic Theory and Applications of High Performance Computing)
  • Feature Paper
  • Article
  • Open Access
4 Citations
3,284 Views
13 Pages

Energy and Precision Evaluation of a Systolic Array Accelerator Using a Quantization Approach for Edge Computing

  • Alejandra Sanchez-Flores,
  • Jordi Fornt,
  • Lluc Alvarez and
  • Bartomeu Alorda-Ladaria

This paper focuses on the implementation of a neural network accelerator optimized for speed and energy efficiency, for use in embedded machine learning. Specifically, we explore power reduction at the hardware level through systolic array and low-pr...

(This article belongs to the Special Issue Recent Advances and Challenges in IoT, Cloud and Edge Coexistence)
  • Article
  • Open Access
18 Citations
5,801 Views
19 Pages

Power-Intent Systolic Array Using Modified Parallel Multiplier for Machine Learning Acceleration

  • Kashif Inayat,
  • Fahad Bin Muslim,
  • Javed Iqbal,
  • Syed Agha Hassnain Mohsan and ,
  • Hend Khalid Alkahtani and
  • Samih M. Mostafa
  • + 1 author+ 0 author

26 April 2023

Systolic arrays are an integral part of many modern machine learning (ML) accelerators due to their efficiency in performing matrix multiplication that is a key primitive in modern ML models. Current state-of-the-art in systolic array-based accelerat...

(This article belongs to the Section Intelligent Sensors)
  • Article
  • Open Access
3 Citations
3,898 Views
12 Pages

14 March 2025

The rapid development of deep neural networks (DNNs), such as convolutional neural networks and transformer-based large language models, has significantly advanced AI applications. However, these advances have introduced substantial computational and...

  • Article
  • Open Access
12 Citations
9,138 Views
11 Pages

Systolic arrays are the primary part of modern deep learning accelerators and are being used widely in real-life applications such as self-driving cars. This paper presents a novel factored systolic array, where the carry propagation adder for accumu...

(This article belongs to the Section Artificial Intelligence Circuits and Systems (AICAS))
  • Article
  • Open Access
786 Views
17 Pages

The increasing computational demands of artificial intelligence (AI) inference at the edge require hardware accelerators capable of overcoming the von Neumann bottleneck while operating under power constraints. Conventional digital architectures base...

(This article belongs to the Section Microelectronics)
  • Article
  • Open Access
5 Citations
3,564 Views
10 Pages

In recent years, deep neural networks (DNNs) have addressed new applications with intelligent autonomy, often achieving higher accuracy than human experts. This capability comes at the expense of the ever-increasing complexity of emerging DNNs, causi...

(This article belongs to the Section Network)
  • Article
  • Open Access
8 Citations
6,873 Views
14 Pages

A key part of modern deep neural network (DNN) applications is matrix multiplication. As DNN applications are becoming more diverse, there is a need for both dense and sparse matrix multiplications to be accelerated by hardware. However, most hardwar...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
1 Citations
2,746 Views
14 Pages

A Systolic Accelerator for Neuromorphic Visual Recognition

  • Shuo Tian,
  • Lei Wang,
  • Shi Xu,
  • Shasha Guo and ,
  • Zhijie Yang,
  • Jianfeng Zhang and
  • Weixia Xu
  • + 2 authors+ 0 author

15 October 2020

Advances in neuroscience have encouraged researchers to focus on developing computational models that behave like the human brain. HMAX is one of the potential biologically inspired models that mimic the primate visual cortex’s functions and st...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
7 Citations
7,382 Views
26 Pages

Deep Learning Accelerators’ Configuration Space Exploration Effect on Performance and Resource Utilization: A Gemmini Case Study

  • Dennis Agyemanh Nana Gookyi,
  • Eunchong Lee,
  • Kyungho Kim,
  • Sung-Joon Jang and
  • Sang-Seol Lee

21 February 2023

Though custom deep learning (DL) hardware accelerators are attractive for making inferences in edge computing devices, their design and implementation remain a challenge. Open-source frameworks exist for exploring DL hardware accelerators. Gemmini is...

(This article belongs to the Section Intelligent Sensors)
  • Article
  • Open Access
32 Citations
7,674 Views
15 Pages

An FPGA-Based LSTM Acceleration Engine for Deep Learning Frameworks

  • Dazhong He,
  • Junhua He,
  • Jun Liu,
  • Jie Yang and ,
  • Qing Yan and
  • Yang Yang
  • + 1 author+ 0 author

Over the past two decades, Long Short-Term Memory (LSTM) networks have been used to solve problems that require modeling of long sequence because they can selectively remember certain patterns over a long period, thus outperforming traditional feed-f...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
1 Citations
848 Views
23 Pages

AI accelerators increasingly operate under tight power, thermal, voltage, and timing margins, making workload-dependent thermal nonuniformity an important reliability concern. In systolic AI accelerators, localized activity concentration can create s...

(This article belongs to the Topic Advanced Integrated Circuit Design and Application)
  • Article
  • Open Access
18 Citations
3,480 Views
19 Pages

Accelerating Strawberry Ripeness Classification Using a Convolution-Based Feature Extractor along with an Edge AI Processor

  • Joungmin Park,
  • Jinyoung Shin,
  • Raehyeong Kim,
  • Seongmo An and ,
  • Sangho Lee,
  • Jinyeol Kim,
  • Jongwon Oh,
  • Youngwoo Jeong,
  • Soohee Kim and
  • Seung Eun Lee
  • + 6 authors+ 1 author

Image analysis-based artificial intelligence (AI) models leveraging convolutional neural networks (CNN) take a significant role in evaluating the ripeness of strawberry, contributing to the maximization of productivity. However, the convolution, whic...

(This article belongs to the Section Artificial Intelligence Circuits and Systems (AICAS))
  • Article
  • Open Access
550 Views
26 Pages

Palm-vein recognition is an attractive biometric modality for secure access control because its subcutaneous vascular patterns are difficult to observe and reproduce externally. However, existing studies optimize the recognition algorithm and the har...

  • Article
  • Open Access
445 Views
17 Pages

Fault-injection campaigns are widely used to evaluate silent data corruption (SDC) in AI hardware, but exhaustive campaigns over workloads, dataflows, processing elements, and datapath roles are expensive. This paper presents an architecture-level ri...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
1 Citations
2,602 Views
18 Pages

On-chip memory is one of the core components of deep learning accelerators. In general, the area used by the on-chip memory accounts for around 30% of the total chip area. With the increasing complexity of deep learning algorithms, it will become a c...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
3,387 Views
18 Pages

Flexible Convolver for Convolutional Neural Networks Deployment onto Hardware-Oriented Applications

  • Moisés Arredondo-Velázquez,
  • Paulo Aaron Aguirre-Álvarez,
  • Alfredo Padilla-Medina,
  • Alejandro Espinosa-Calderon and ,
  • Juan Prado-Olivarez and
  • Javier Diaz-Carmona
  • + 1 author+ 0 author

21 December 2022

This paper introduces a flexible convolver capable of adapting to the different convolution layer configurations of state-of-the-art Convolution Neural Networks (CNNs). The use of two proposed programmable components achieves this adaptability. A Pro...

(This article belongs to the Section Electrical, Electronics and Communications Engineering)
  • Article
  • Open Access
3 Citations
3,327 Views
19 Pages

29 February 2024

This paper conducts a comprehensive study on intermittent computing within IoT environments, emphasizing the interplay between different dataflows—row, weight, and output—and a variety of non-volatile memory technologies. We then delve in...

(This article belongs to the Section E: Engineering and Technology)
  • Article
  • Open Access
12 Citations
15,376 Views
14 Pages

Efficient Systolic-Array Redundancy Architecture for Offline/Online Repair

  • Keewon Cho,
  • Ingeol Lee,
  • Hyeonchan Lim and
  • Sungho Kang

Neural-network computing has revolutionized the field of machine learning. The systolic-array architecture is a widely used architecture for neural-network computing acceleration that was adopted by Google in its Tensor Processing Unit (TPU). To ensu...

(This article belongs to the Section Circuit and Signal Processing)
  • Article
  • Open Access
2,582 Views
14 Pages

Hardware Acceleration of Identifying Barcodes in Multiplexed Nanopore Sequencing

  • Wenjie Hu,
  • Yuxin Zhang,
  • Hongrui Zhang and
  • Weigang Chen

19 August 2022

In multiplexed sequencing, the identification of DNA sequencing barcodes can effectively reduce the probability of sample misassignment. However, the great quantity of sequence data requires a high-throughput identification method. Therefore, based o...

(This article belongs to the Section Bioelectronics)
  • Article
  • Open Access
8 Citations
6,354 Views
18 Pages

High-Speed CNN Accelerator SoC Design Based on a Flexible Diagonal Cyclic Array

  • Dong-Yeong Lee,
  • Hayotjon Aliev,
  • Muhammad Junaid,
  • Sang-Bo Park and ,
  • Hyung-Won Kim,
  • Keon-Myung Lee and
  • Sang-Hoon Sim
  • + 2 authors+ 0 author

The latest convolutional neural network (CNN) models for object detection include complex layered connections to process inference data. Each layer utilizes different types of kernel modes, so the hardware needs to support all kernel modes at an opti...

(This article belongs to the Special Issue CMOS Integrated Circuits Design)
  • Article
  • Open Access
3 Citations
4,001 Views
12 Pages

27 August 2022

In this paper, a small-area low-power 64-bit integer multiplier is presented, which is suitable for portable devices or wireless applications. To save the area cost and power consumption, an input vector systolic (IVS) structure is proposed based on...

(This article belongs to the Special Issue VLSI Circuits & Systems Design)
  • Article
  • Open Access
2 Citations
3,740 Views
14 Pages

As one of the most important and computationally-intensive parts in bioinformatics analysis, the Pair Hidden Markov Model (Pair-HMM) forward algorithm is widely recognized and has great potential. Therefore, it is important to accelerate the process...

(This article belongs to the Section Computer Science & Engineering)
  • Article
  • Open Access
8 Citations
5,372 Views
16 Pages

An Energy-Efficient Convolutional Neural Network Processor Architecture Based on a Systolic Array

  • Chen Zhang,
  • Xin’an Wang,
  • Shanshan Yong,
  • Yining Zhang and ,
  • Qiuping Li and
  • Chenyang Wang
  • + 1 author+ 0 author

9 December 2022

Deep convolutional neural networks (CNNs) have shown strong abilities in the application of artificial intelligence. However, due to their extensive amount of computation, traditional processors have low energy efficiency when executing CNN algorithm...

  • Article
  • Open Access
13 Citations
11,056 Views
23 Pages

Convolutional neural networks (CNNs) have demonstrated promising results in various applications such as computer vision, speech recognition, and natural language processing. One of the key computations in many CNN applications is matrix multiplicati...

(This article belongs to the Section Circuit and Signal Processing)
  • Article
  • Open Access
6 Citations
4,040 Views
20 Pages

Implementation of Autoencoders with Systolic Arrays through OpenCL

  • Rafael Gadea-Gironés,
  • Vicente Herrero-Bosch,
  • Jose Monzó-Ferrer and
  • Ricardo Colom-Palero

In the world of algorithm acceleration and the implementation of deep neural networks’ recall phase, OpenCL based solutions have a clear tendency to produce perfectly adapted kernels in graphic processor unit (GPU) architectures. However, they...

(This article belongs to the Special Issue Advanced AI Hardware Designs Based on FPGAs)
  • Article
  • Open Access
723 Views
13 Pages

TR-ABFT: Tile-Resilient Fault Detection for Neural Processing Units

  • Yang Hua,
  • Yunhong Bai,
  • Bo Wang,
  • Wei Zhuang and
  • Yuanfu Zhao

Spaceborne neural processing units (NPUs) increasingly support real-time deep-learning inference, but their dense multiply-accumulate arrays are vulnerable to radiation-induced soft errors. Conventional radiation-hardening methods improve reliability...

(This article belongs to the Special Issue Artificial Intelligence and Microsystems)
  • Article
  • Open Access
9 Citations
5,655 Views
19 Pages

Challenges and Opportunities in Near-Threshold DNN Accelerators around Timing Errors

  • Pramesh Pandey,
  • Noel Daniel Gundi,
  • Prabal Basu,
  • Tahmoures Shabanian and ,
  • Mitchell Craig Patrick,
  • Koushik Chakraborty and
  • Sanghamitra Roy
  • + 2 authors+ 0 author

AI evolution is accelerating and Deep Neural Network (DNN) inference accelerators are at the forefront of ad hoc architectures that are evolving to support the immense throughput required for AI computation. However, much more energy efficient design...

(This article belongs to the Special Issue Circuits and Systems Advances in Near Threshold Computing)
  • Article
  • Open Access
22 Citations
6,219 Views
20 Pages

Efficient FPGA Implementation of Convolutional Neural Networks and Long Short-Term Memory for Radar Emitter Signal Recognition

  • Bin Wu,
  • Xinyu Wu,
  • Peng Li,
  • Youbing Gao and ,
  • Jiangbo Si and
  • Naofal Al-Dhahir
  • + 1 author+ 0 author

30 January 2024

In recent years, radar emitter signal recognition has enjoyed a wide range of applications in electronic support measure systems and communication security. More and more deep learning algorithms have been used to improve the recognition accuracy of...

(This article belongs to the Special Issue Radar Signal Detection, Recognition and Identification)
  • Article
  • Open Access
3 Citations
3,202 Views
23 Pages

Voltage Scaled Low Power DNN Accelerator Design on Reconfigurable Platform

  • Rourab Paul,
  • Sreetama Sarkar,
  • Suman Sau,
  • Sanghamitra Roy and ,
  • Koushik Chakraborty and
  • Amlan Chakrabarti
  • + 1 author+ 0 author

The exponential emergence of Field-Programmable Gate Arrays (FPGAs) has accelerated research on hardware implementation of Deep Neural Networks (DNNs). Among all DNN processors, domain-specific architectures such as Google’s Tensor Processor Un...

(This article belongs to the Special Issue Embedded Systems for Neural Network Applications)
  • Article
  • Open Access
723 Views
22 Pages

Hardware Accelerator Design for MUSIC-DOA Estimation with Bilateral Jacobi Optimization

  • Yafan Gao,
  • Weijiang Wang,
  • Chengbo Xue,
  • Shiwei Ren and ,
  • Kuanhao Liu and
  • Xiangnan Li
  • + 1 author+ 0 author

Real-time Direction of Arrival (DOA) estimation demands high computational throughput and numerical precision. Consequently, dedicated hardware accelerators are essential. This paper presents an architecture to accelerate the MUSIC algorithm using an...

(This article belongs to the Special Issue New Advances of FPGAs in Signal Processing)
  • Article
  • Open Access
9 Citations
6,437 Views
18 Pages

Accelerating Binary String Comparisons with a Scalable, Streaming-Based System Architecture Based on FPGAs

  • Sarah Pilz,
  • Florian Porrmann,
  • Martin Kaiser,
  • Jens Hagemeyer and ,
  • James M. Hogan and
  • Ulrich Rückert
  • + 1 author+ 0 author

21 February 2020

This paper is concerned with Field Programmable Gate Arrays (FPGA)-based systems for energy-efficient high-throughput string comparison. Modern applications which involve comparisons across large data sets—such as large sequence sets in molecul...

  • Article
  • Open Access
8 Citations
5,704 Views
27 Pages

A Low Complexity, High Throughput DoA Estimation Chip Design for Adaptive Beamforming

  • Kuan-Ting Chen,
  • Wei-Hsuan Ma,
  • Yin-Tsung Hwang and
  • Kuan-Ying Chang

Direction of Arrival (DoA) estimation is essential to adaptive beamforming widely used in many radar and wireless communication systems. Although many estimation algorithms have been investigated, most of them focus on the performance enhancement asp...

(This article belongs to the Special Issue Advanced Integrated Circuits Technology)
  • Feature Paper
  • Article
  • Open Access
3 Citations
2,358 Views
18 Pages

30 December 2024

Graph-based neural networks have proven to be useful in molecular property prediction, a critical component of computer-aided drug discovery. In this application, in response to the growing demand for improved computational efficiency and localized e...

(This article belongs to the Special Issue Machine Learning in Electronic and Biomedical Engineering, 3rd Edition)
  • Article
  • Open Access
4,263 Views
18 Pages

Understanding Timing Error Characteristics from Overclocked Systolic Multiply–Accumulate Arrays in FPGAs

  • Andrew Chamberlin,
  • Andrew Gerber,
  • Mason Palmer,
  • Tim Goodale and ,
  • Noel Daniel Gundi,
  • Koushik Chakraborty and
  • Sanghamitra Roy
  • + 2 authors+ 0 author

Artificial Intelligence (AI) hardware accelerators have seen tremendous developments in recent years due to the rapid growth of AI in multiple fields. Many such accelerators comprise a Systolic Multiply–Accumulate Array (SMA) as its computation...

  • Article
  • Open Access
861 Views
33 Pages

Asynchronous Co-Execution of PyTorch on Zynq-7000: FPGA Matrix Delegation and PS–PL Overlap for End-to-End Inference Throughput

  • Omar Hernandez-Yañez,
  • Alejandro Juarez-Lora,
  • Jesús Yalja Montiel-Pérez,
  • Victor H. Ponce-Ponce and
  • Heron Molina-Lozano

Embedded systems increasingly require on-device deep learning, yet their processors must simultaneously handle real-time sensing, networking administration, and data control. Existing Field-Programmable Gate Array (FPGA) accelerators typically target...

(This article belongs to the Special Issue Hardware Acceleration for Machine Learning, 2nd Edition)
  • Article
  • Open Access
1 Citations
1,580 Views
28 Pages

Edge Detection on a 2D-Mesh NoC with Systolic Arrays: From FPGA Validation to GDSII Proof-of-Concept

  • Emma Mascorro-Guardado,
  • Susana Ortega-Cisneros,
  • Francisco Javier Ibarra-Villegas,
  • Jorge Rivera and ,
  • Héctor Emmanuel Muñoz-Zapata and
  • Emilio Isaac Baungarten-Leon
  • + 1 author+ 0 author

9 January 2026

Edge detection is a key building block in real-time image-processing applications such as drone-based infrastructure inspection, autonomous navigation, and remote sensing. However, its computational cost remains a challenge for resource-constrained e...

(This article belongs to the Special Issue Advanced Integrated Circuit Design and Applications)
  • Article
  • Open Access
4 Citations
3,415 Views
23 Pages

A Heterogeneous Inference Framework for a Deep Neural Network

  • Rafael Gadea-Gironés,
  • José Luís Rocabado-Rocha,
  • Jorge Fe and
  • Jose M. Monzo

Artificial intelligence (AI) is one of the most promising technologies based on machine learning algorithms. In this paper, we propose a workflow for the implementation of deep neural networks. This workflow attempts to combine the flexibility of hig...

(This article belongs to the Special Issue Advanced Machine Learning, Pattern Recognition, and Deep Learning Technologies: Methodologies and Applications)