Hardware Perspective Research Articles

Characterizing and Understanding HGNN Training on GPUs

Owing to their remarkable representation capabilities for heterogeneous graph data, Heterogeneous Graph Neural Networks (HGNNs) have been widely adopted in many critical real-world domains such as recommendation systems and medical analysis. Prior to their practical application, identifying the optimal HGNN model parameters tailored to specific tasks through extensive training is a time-consuming and costly process. To enhance the efficiency of HGNN training, it is essential to characterize and analyze the execution semantics and patterns within the training process to identify performance bottlenecks. In this study, we conduct a comprehensive quantification and in-depth analysis of two mainstream HGNN training scenarios, including single-GPU and multi-GPU distributed training. Based on the characterization results, we reveal the performance bottlenecks and their underlying causes in different HGNN training scenarios and propose optimization guidelines from both software and hardware perspectives.

Artificial Skin Based on Visuo‐Tactile Sensing for 3D Shape Reconstruction: Material, Method, and Evaluation

AbstractArtificial skin has shown great potential in robot perception and human healthcare. It provides multifunctional tactile sensing, including 3D shape reconstruction, contact feedback, and temperature perception, where the 3D reconstruction function is indispensable for dexterous hands in tactile cognition and interaction. Vision‐based tactile sensor (VTS) is an innovative bionic tactile sensor and supports high‐resolution, high‐precision, and high‐density tactile reconstruction compared with electronic tactile sensors. Considering the unique contribution of visuo‐tactile sensing to artificial skin, this review focuses on the 3D reconstruction techniques of the VTS. 3D reconstruction methods are classified into five categories based on sensing modalities, hardware categories, and modeling approaches: 1) photometric stereo, 2) binocular depth calibration, 3) optical flow, 4) deep learning, and 5) ToF (time of flight). In addition, the association and difference of reconstruction methods are analyzed from the hardware perspective, and the development and technological details of 3D reconstruction are summarized. On this basis, the challenges and development direction are discussed. This review can be viewed as a technology guide to provide references for interested researchers. Furthermore, it is expected to promote the extensive application of the VTS in artificial skins.

Design of an FPGA-based short-circuit fault diagnosis algorithm for DC current-limiting fuses

Abstract In recent years, DC current-limiting fuses have played a crucial role in the comprehensive power systems of ships, rail transport, and new energy sectors, ensuring the safe and reliable operation of DC power systems. However, with the increase in the capacity of DC power systems, short-circuit faults can lead to a rise in short-circuit current at rates of up to 40 A/μs, with peak short-circuit currents reaching up to 105KA. This poses higher demands on the accuracy and speed of fuse diagnostics to interrupt short-circuit currents. To address this issue, this paper introduces a fault diagnostic algorithm based on FPGA that can calculate the rise rate of short-circuit currents in real-time: employing FPGA with its high integration, abundant logic resources, and parallel data processing capabilities as the core of the measurement and control system to enhance the speed of short-circuit fault diagnosis from a hardware perspective; replacing the common large current trip protection and DDL protection algorithms with the Newton interpolation algorithm. This improvement calculates the rise rate of each current data point instead of the average rise rate over a period, thus enhancing the speed and accuracy of short-circuit fault diagnosis from a software perspective. The paper also discusses the design of the system’s hardware and software and the construction of the experimental platform. Simulation and experimental results show that the fault diagnosis time of the algorithm has been reduced from 98μs to 36μs, with the numerical error in the short-circuit current rise rate within 2%, and a significant reduction in the peak value of short-circuit current, meeting performance requirements and effectively ensuring the safe and reliable operation of DC power systems.

A Systematic Review of Fast, Scalable and Efficient Hardware Implementations of Elliptic Curve Cryptography for Blockchain

Blockchain technology entered the enterprise domain under the name of permissioned blockchains and hybrid or verifiable database systems, as they provide a distributed solution that allows multiple distrusting parties to share common information. One drawback of these systems is the overhead added by the cryptographic functions which impacts the throughput in terms of transactions per second and increases the latency of transaction processing. Many of the cryptographic functions and protocols used in blockchains are based on Elliptic Curve Cryptography (ECC). Unfortunately, ECC operations such as modulo inverse or scalar point multiplication have considerable latency which causes the slowdown of the entire system. In such situations, reconfigurable computing architectures, such as FPGAs, can be used to offload these tasks to overcome the performance loss. This survey analyzes the current state-of-the-art designs and implementations of ECC from a hardware perspective. We use a PRISMA-based approach to filter recent publications and to reduce their number from over 16,000 to only 43 highly relevant designs. In the end, we show that very few designs are able to fulfill all three properties of high performance, scalability, and efficiency.

Smart Classrooms: How Sensors and AI Are Shaping Educational Paradigms.

The integration of advanced technologies is revolutionizing classrooms, significantly enhancing their intelligence, interactivity, and personalization. Central to this transformation are sensor technologies, which play pivotal roles. While numerous surveys summarize research progress in classrooms, few studies focus on the integration of sensor and AI technologies in developing smart classrooms. This systematic review classifies sensors used in smart classrooms and explores their current applications from both hardware and software perspectives. It delineates how different sensors enhance educational outcomes and the crucial role AI technologies play. The review highlights how sensor technology improves the physical classroom environment, monitors physiological and behavioral data, and is widely used to boost student engagements, manage attendance, and provide personalized learning experiences. Additionally, it shows that combining sensor software algorithms with AI technology not only enhances the data processing and analysis efficiency but also expands sensor capabilities, enriching their role in smart classrooms. The article also addresses challenges such as data privacy protection, cost, and algorithm optimization associated with emerging sensor technologies, proposing future research directions to advance educational sensor technologies.

SUPER-MAN: SUPERnumerary robotic bodies for physical assistance in huMAN–robot conjoined actions

This paper presents a mobile supernumerary robotic approach to physical assistance in human–robot conjoined actions. The study starts with the description of the SUPER-MAN concept. The idea is to develop and utilize mobile collaborative systems that can follow human loco-manipulation commands to perform industrial tasks through three main components: (i) an admittance-type interface, (ii) a human–robot interaction controller and (iii) a supernumerary robotic body. Next, we present two possible implementations within the framework — from theoretical and hardware perspectives. The first system is called MOCA-MAN, and is composed of a redundant torque-controlled robotic arm and an omni-directional mobile platform. The second one is called Kairos-MAN, formed by a high-payload 6-DoF velocity-controlled robotic arm and an omni-directional mobile platform. The systems share the same admittance interface, through which user wrenches are translated to loco-manipulation commands, generated by whole-body controllers of each system. Besides, a thorough user-study with multiple and cross-gender subjects is presented to reveal the quantitative performance of the two systems in effort demanding and dexterous tasks. Moreover, we provide qualitative results from the NASA-TLX questionnaire to demonstrate the SUPER-MAN approach’s potential and its acceptability from the users’ viewpoint.

Common sleep data pipeline for combined data sets.

Over the past few years, sleep research has shown impressive performance of deep neural networks in the area of automatic sleep-staging. Recent studies have demonstrated the necessity of combining multiple data sets to obtain sufficiently generalizing results. However, working with large amounts of sleep data can be challenging, both from a hardware perspective and because of the different preprocessing steps necessary for distinct data sources. Here we review the possible obstacles and present an open-source pipeline for automatic data loading. Our solution includes both a standardized data store as well as a 'data serving' portion which can be used to train neural networks on the standardized data, allowing for different configuration options for different studies and machine learning designs. The pipeline, including implementation, is made public to ensure better and more reproducible sleep research.

Development of a Premium Tea-Picking Robot Incorporating Deep Learning and Computer Vision for Leaf Detection

Premium tea holds a significant place in Chinese tea culture, enjoying immense popularity among domestic consumers and an esteemed reputation in the international market, thereby significantly impacting the Chinese economy. To tackle challenges associated with the labor-intensive and inefficient manual picking process of premium tea, and to elevate the competitiveness of the premium tea sector, our research team has developed and rigorously tested a premium tea-picking robot that harnesses deep learning and computer vision for precise leaf recognition. This innovative technology has been patented by the China National Intellectual Property Administration (ZL202111236676.7). In our study, we constructed a deep-learning model that, through comprehensive data training, enabled the robot to accurately recognize tea buds. By integrating computer vision techniques, we achieved exact positioning of the tea buds. From a hardware perspective, we employed a high-performance robotic arm to ensure stable and efficient picking operations even in complex environments. During the experimental phase, we conducted detailed validations on the practical application of the YOLOv8 algorithm in tea bud identification. When compared to the YOLOv5 algorithm, YOLOv8 exhibited superior accuracy and reliability. Furthermore, we performed comprehensive testing on the path planning for the picking robotic arm, evaluating various algorithms to determine the most effective path planning approach for the picking process. Ultimately, we conducted field tests to assess the robot’s performance. The results indicated a 62.02% success rate for the entire picking process of the premium tea-picking robot, with an average picking time of approximately 1.86 s per qualified tea bud. This study provides a solid foundation for further research, development, and deployment of premium tea-picking robots, serving as a valuable reference for the design of other crop-picking robots as well.

Reliability Assessment of Wireless Sensor Networks by Strain-Based Region Analysis for Redundancy Estimation in Measurements on the Example of an Aircraft Wing Box.

Wireless sensor networks (WSNs) are attracting increasing research interest due to their ability to monitor large areas independently. Their reliability is a crucial issue, as it is influenced by hardware, data, and energy-related factors such as loading conditions, signal attenuation, and battery lifetime. Proper selection of sensor node positions is essential to maximise system reliability during the development of products equipped with WSNs. For this purpose, this paper presents an approach to estimate WSN system reliability during the development phase based on the analysis of measurements, using strain measurements in finite element (FE) models as an example. The approach involves dividing the part under consideration into regions with similar strains using a region growing algorithm (RGA). The WSN configuration is then analysed for reliability based on data paths and measurement redundancy resulting from the sensor positions in the identified measuring regions. This methodology was tested on an exemplary WSN configuration at an aircraft wing box under bending load and found to effectively estimate the hardware perspective on system reliability. Therefore, the methodology and algorithm show potential for optimising sensor node positions to achieve better reliability results.

Enhancing the performance of coherent Ising machines in the large-noise regime with a fifth-order nonlinearity

Coherent Ising machines (CIMs), leveraging the bistable physical properties of coherent light to emulate Ising spins, exhibit great potential as hardware accelerators for tackling complex combinatorial optimization problems. Recent advances have demonstrated that the performance of CIMs can be enhanced either by incorporating large random noise or higher-order nonlinearities, yet their combined effects on CIM performance remain mainly unexplored. In this work, we develop a numerical CIM model that utilizes a tunable fifth-order polynomial nonlinear dynamic function under large noise levels, which has the potential to be implemented in all-optical platforms. We propose a normal form of a CIM model that allows for both supercritical and subcritical pitchfork bifurcation operational regimes, with fifth-order nonlinearity and tunable hyperparameters to control the Ising spin dynamics. In the benchmark studies, we simulate various sets of MaxCut problems using our fifth-order polynomial CIM model. The results show a significant performance improvement, achieving an average of 59.5% improvement in median time-to-solution (TTS) and an average of 6 times improvement in median success rate (SR) for dense Maxcut problems in the BiqMac library, compared to the commonly used third-order polynomial CIM model with low noise. The fifth-order polynomial CIM model in the large-noise regime also shows better performance trends as the problem size scales up. These findings reveal the enhancements on the computational performance of Ising machines in the large-nose regime from fifth-order nonlinearity, showing important implications for both simulation and hardware perspectives.

A Survey on Graph Neural Network Acceleration: A Hardware Perspective

A Survey on Graph Neural Network Acceleration: A Hardware Perspective

Noctua2 Supercomputer

Noctua 2 is a supercomputer operated at the Paderborn Center for Parallel Computing (PC2) at Paderborn University in Germany. Noctua 2 was inaugurated in 2022 and is an Atos BullSequana XH2000 system. It consists mainly of three node types: 1) CPU Compute nodes with AMD EPYC processors in different main memory configurations, 2) GPU nodes with NVIDIA A100 GPUs, and 3) FPGA nodes with Xilinx Alveo U280 and Intel Stratix 10 FPGA cards. While CPUs and GPUs are known off-the-shelf components in HPC systems, the operation of a large number of FPGA cards from different vendors and a dedicated FPGA-to-FPGA network are unique characteristics of Noctua 2. This paper describes in detail the overall setup of Noctua 2 and gives insights into the operation of the cluster from a hardware, software and facility perspective.

SQUID magnetoneurography: an old-fashioned yet new tool for noninvasive functional imaging of spinal cords and peripheral nerves.

We are engaged in the development and clinical application of a neural magnetic field measurement system that utilizes biomagnetic measurements to observe the activity of the spinal cord and peripheral nerves. Unlike conventional surface potential measurements, biomagnetic measurements are not affected by the conductivity distribution within the body, making them less influenced by the anatomical structure of body tissues. Consequently, functional testing using biomagnetic measurements can achieve higher spatial resolution compared to surface potential measurements. The neural magnetic field measurement, referred to as magnetoneurography, takes advantage of these benefits to enable functional testing of the spinal cord and peripheral nerves, while maintaining high spatial resolution and noninvasiveness. Our magnetoneurograph system is based on superconducting quantum interference devices (SQUIDs) similar to the conventional biomagnetic measurement systems. Various design considerations have been incorporated into the SQUID sensor array structure and signal processing software to make it suitable for detecting neural signal propagation along spinal cord and peripheral nerve. The technical validation of this system began in 1999 with a 3-channel SQUID system. Over the course of more than 20 years, we have continued technological development through medical-engineering collaboration, and in the latest prototype released in 2020, neural function imaging of the spinal cord and peripheral nerves, which could also be applied for the diagnosis of neurological disorders, has become possible. This paper provides an overview of the technical aspects of the magnetoneurograph system, covering the measurement hardware and software perspectives for providing diagnostic information, and its applications. Additionally, we discuss the integration with a helium recondensing system, which is a key factor in reducing running costs and achieving practicality in hospitals.

Novel Approach for Network Data Exchange Among Safety-Related Control Systems of Nuclear Reactors

Industrially mature digital networking technologies form the backbone of the distributed control architectures of nuclear reactors. Typically, hundreds of control systems responsible for assuring safer and secure plant operations are located in the field and interconnected via wired isolated local area networks, which are also connected to centralized servers and operating consoles located in the main control room. Many of such control systems require data processed by other systems situated far away in the plant, which mandates multiple concurrent reliable TCP sessions to be active or live for such field-level control systems. Multi-socket communication beyond 10 is practically difficult for embedded systems with limited memory and processing capabilities; hence, data accumulation, extraction, and routing are typically done with the help of commercially available high-performance servers and switches. The servers and operating consoles of a plant are generally Linux-based commercial-off-the-shelf (COTS) industrial computers, whose development lifecycle is not accessible for review and in accordance with safety guides; hence, qualifying such components for safety operations is extremely difficult. This paper presents a novel way of independently offloading safety-related data transactions from such COTS components (reducing dependency) with the help of very simple, reliable, and secure hardware, which is referred to as a data exchange and processing unit (DEPU). An approach is proposed to realize distributed control architecture with this proposed hardware (to simplify qualification tasks) along with existing servers and operating consoles without compromising powerful and convenient human-machine interface as well as the data logging capabilities of such components. Hardware and software design perspectives, performance characterization, and qualification testing of the DEPU in accordance with nuclear power plant instrumentation and control system design guidelines are also briefed in this paper.

Practical solutions in fully homomorphic encryption: a survey analyzing existing acceleration methods

Fully homomorphic encryption (FHE) has experienced significant development and continuous breakthroughs in theory, enabling its widespread application in various fields, like outsourcing computation and secure multi-party computing, in order to preserve privacy. Nonetheless, the application of FHE is constrained by its substantial computing overhead and storage cost. Researchers have proposed practical acceleration solutions to address these issues. This paper aims to provide a comprehensive survey for systematically comparing and analyzing the strengths and weaknesses of FHE acceleration schemes, which is currently lacking in the literature. The relevant researches conducted between 2019 and 2022 are investigated. We first provide a comprehensive summary of the latest research findings on accelerating FHE, aiming to offer valuable insights for researchers interested in FHE acceleration. Secondly, we classify existing acceleration schemes from algorithmic and hardware perspectives. We also propose evaluation metrics and conduct a detailed comparison of various methods. Finally, our study presents the future research directions of FHE acceleration, and also offers both guidance and support for practical application and theoretical research in this field.

Does hyperspectral always matter? A critical assessment of near infrared versus hyperspectral near infrared in the study of heterogeneous samples

Near Infrared spectroscopy (NIR), in combination with Chemometrics, has been used for many years in diverse scenarios, mostly focused on the classification and quantitation of properties in food, pharmaceutical preparations, artwork material, etc. This success has been possible due to their desirable properties: fast, reliable (under certain conditions), non-destructive, easy to implement from a hardware perspective, and able to create robust and transferable multivariate models.For some years now, another modality has been gaining the attention of NIR users, especially in the Food sector. That is the plausibility of using NIR in the hyperspectral (HSI) domain. This adds to the previously mentioned abilities, the benefit of scanning the whole surface of samples, acquiring much richer spatial information and, therefore, assuring the quality of the final product more accurately by including parameters that depend on the surface distribution of certain components. This is especially relevant in heterogeneous samples. While this statement is generally true, there are certain situations where this oversampling feature is not strictly needed, and the problem can be easily solved with a classical NIR spectrophotometer. Besides, NIR-hyperspectral imaging (NIR-HSI), despite the abovementioned advantages, has several drawbacks that must be highlighted as well, like their measuring speed, instability, or price.This manuscript will demonstrate that for certain situations, tuning the focal distance of a NIR spectrophotometer is a more feasible, reliable, and inexpensive strategy to collect all the needed information of samples with a certain degree of heterogeneity.

A novel high-frequency resonance controllable pneumatic actuator and its high-precision motion trajectory tracking control

The strong nonlinearity and uncertainty of the pneumatic cylinder friction is one of the main reasons for the poor control accuracy of the pneumatic cylinder. For this reason, improving the friction characteristics of the pneumatic cylinder should contribute to the improvement of the pneumatic cylinder motion trajectory tracking accuracy. Reducing friction is a common method to improve the friction characteristics of pneumatic cylinders. In this paper, a novel longitudinal resonance friction-reducing pneumatic cylinder based on the principle of inverse piezoelectric effect is developed from the hardware perspective. Meanwhile, a promising adaptive robust controller (ARC) with online parameter identification and nonlinear robust control function is designed for the pneumatic position servo system from the software perspective. Finally, this paper focuses on exploring the possibility of further improving the accuracy by excitation of high-frequency resonance of the developed novel pneumatic actuator on the basis of the accuracy achievable at the software level (control algorithm). Experimental results show that friction reduction by means of high-frequency resonance can significantly improve the motion trajectory tracking accuracy of the pneumatic system. The maximum tracking error is reduced by 17.95% and 27.87% when tracking sinusoidal reference trajectories of 0.5 Hz and 0.25 Hz, respectively. The maximum tracking error is reduced by 15.79% when tracking a smooth square reference trajectory of 0.25 Hz. In addition, the designed controller is effective and robust to parameter variations and sudden perturbations.

Evolution and impact of high content imaging.

The field of high content imaging has steadily evolved and expanded substantially across many industry and academic research institutions since it was first described in the early 1990's. High content imaging refers to the automated acquisition and analysis of microscopic images from a variety of biological sample types. Integration of high content imaging microscopes with multiwell plate handling robotics enables high content imaging to be performed at scale and support medium- to high-throughput screening of pharmacological, genetic and diverse environmental perturbations upon complex biological systems ranging from 2D cell cultures to 3D tissue organoids to small model organisms. In this perspective article the authors provide a collective view on the following key discussion points relevant to the evolution of high content imaging: • Evolution and impact of high content imaging: An academic perspective • Evolution and impact of high content imaging: An industry perspective • Evolution of high content image analysis • Evolution of high content data analysis pipelines towards multiparametric and phenotypic profiling applications • The role of data integration and multiomics • The role and evolution of image data repositories and sharing standards • Future perspective of high content imaging hardware and software.

Performance Analysis and Optimal Design of BATS Code: A Hardware Perspective

Batched Sparse codes (BATS codes) are a class of <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">linear network coding</b> schemes that increase network throughput by converting a multi-hop problem into an end-to-end problem through network coding. It plays a crucial role in future wireless communication, where packet loss is inevitable. This is an enduring problem in many applications, including vehicular networks. While the theory of BATS codes has been developed over the past decade, little research has been done on its hardware implementation, which is important for practical adoption. This paper provides a systematic way to analyze the performance of a BATS hardware accelerator when the code varies. A roofline model which provides an upper bound of the performance considering both computational and input/output constraints is developed. Next, we build a model connecting the BATS code design space with the hardware execution time, which can be used to determine an optimal code. Finally, we propose and test a flexible and scalable BATS code design paradigm for hardware accelerators with numerical results. The same framework could easily be extended to other linear codes and different computing systems.

A Comparative Survey of Instance Selection Methods applied to Non-Neural and Transformer-Based Text Classification

Progress in natural language processing has been dictated by the rule of more : more data, more computing power, more complexity, best exemplified by deep learning Transformers. However, training (or fine-tuning) large dense models for specific applications usually requires significant amounts of computing resources. One way to ameliorate this problem is through data engineering instead of the algorithmic or hardware perspectives. Our focus here is an under-investigated data engineering technique, with enormous potential in the current scenario – Instance Selection (IS) (a.k.a. Selective Sampling, Prototype Selection). The IS goal is to reduce the training set size by removing noisy or redundant instances while maintaining or improving the effectiveness (accuracy) of the trained models and reducing the training process cost. We survey classical and recent state-of-the-art IS techniques and provide a scientifically sound comparison of IS methods applied to an essential natural language processing task—Automatic Text Classification (ATC). IS methods have been normally applied to small tabular datasets and have not been systematically compared in ATC. We consider several neural and non-neural state-of-the-art ATC solutions and many datasets. We answer several research questions based on tradeoffs induced by a tripod (training set reduction, effectiveness, and efficiency). Our answers reveal an enormous unfulfilled potential for IS solutions. Specially, we show that in 12 out of 19 datasets, specific IS methods—namely, Condensed Nearest Neighbor, Local Set-based Smoother, and Local Set Border Selector—can reduce the size of the training set without effectiveness losses. Furthermore, in the case of fine-tuning the Transformer methods, the IS methods reduce the amount of data needed, without losing effectiveness and with considerable training-time gains.