Article
Engineering, Electrical & Electronic
Subhabrata Roy, Abhijit Chandra
Summary: This paper focuses on designing a narrow transition-band FIR filter using a deep learning-based approach, which offers a unified design framework for various FIR filters. Simulation and hardware implementation results demonstrate the effectiveness and advantages of the proposed method.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Electrical & Electronic
Jennifer Hasler, Sahil Shah
Summary: This study investigates a Continuous-Time (CT) Ladder filter implemented on a large-scale Field Programmable Analog Array (FPAA) and characterized on an SoC FPAA. Experimental results demonstrate a reprogrammable CT Analog linear-phase filter by utilizing the ladder filter delay element, making traditionally difficult analog signal processing possible. Additionally, Vector-Matrix Multiplications (VMMs) are shown to compensate for non-idealities in the ladder filter delay-line operation.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2021)
Article
Engineering, Electrical & Electronic
Priyankkumar H. Prajapati, Anand D. Darji
Summary: Wearable/portable electrocardiogram (ECG) devices are beneficial for continuous heart monitoring and early detection of cardiovascular diseases (CVDs). This research proposes a new method using adaptive filtering and moving average filtering to improve ECG signal quality. The hardware design outperforms existing methods in terms of power consumption and signal quality.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Hardware & Architecture
Marcos Cervetto, Edgardo Marchi, Cecilia G. Galarza
Summary: Research and development of algorithms for processing impulse radio ultrawideband signals is a trending issue. A SoC-based platform has been designed for scientific experimentation, achieving a configurable UWB-capable sampling rate through off-the-shelf components.
IEEE EMBEDDED SYSTEMS LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Alexandro Ortiz, Efrain Mendez, David Balderas, Pedro Ponce, Israel Macias, Arturo Molina
Summary: This study describes the implementation of metaheuristic optimization algorithms in hardware and compares five important algorithms. The results demonstrate the feasibility of NI FPGA hardware and reveal differences in device utilization and execution time among the algorithms.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Hardware & Architecture
Aswini K. Samantaray, Pranose J. Edavoor, Amol D. Rahulkar
Summary: This article presents the first dyadic Gabor wavelet filter bank (DGWFB) based on a slight alteration in the orientation parameter without disturbing the remaining Gabor wavelet parameters. A separable VLSI architecture for the proposed DGWFB is introduced. Experimental results show that the proposed DGWFB achieves better performance in medical image retrieval application with significantly reduced digital hardware and processing time compared to existing Gabor wavelet FBs.
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
(2023)
Article
Multidisciplinary Sciences
Ninnart Fuengfusin, Hakaru Tamukoh
Summary: This study introduced a mixed-precision weights network (MPWN) that utilizes three different weight spaces and further developed it from both software and hardware aspects. The software evaluation systematically combined accuracy, sparsity, and number of bits to efficiently search for weight space combinations, while the hardware aspect explored XOR signed-bits implementation for efficient multiplication of weight spaces. The hardware implementation of MPWN in field-programmable gate array showed significant reduction in hardware resources usage and latency compared to a conventional 32-bit floating-point model.
Article
Computer Science, Information Systems
German Cano-Quiveu, Paulino Ruiz-de-clavijo-Vazquez, Manuel J. Bellido, Jorge Juan-Chico, Julian Viejo-Cortes, David Guerrero-Martos, Enrique Ostua-Aranguena
Summary: The paper introduces a hardware-based security framework for IoT devices (E-LUKS) similar to the LUKS solution used in Linux systems, which extends LUKS capabilities by adding integrity and authentication methods, making it a great alternative for providing Full Disk Encryption (FDE) and authentication to a wide range of IoT devices.
Article
Automation & Control Systems
Dhruv M. Patel, Ankit K. Shah
Summary: The article discusses the limitations of existing PLCs based on microprocessors or microcontrollers in industrial applications and introduces a new FPGA-based PLC multi-channel High Speed Counter (HSC) module. The proposed module shows improved performance and flexibility compared to existing PLC-HSC modules, with enhanced accuracy and faster scanning time.
Article
Geochemistry & Geophysics
Daniel Bascones, Carlos Gonzalez, Daniel Mozos
Summary: This article presents a real-time implementation of a hyperspectral image compression algorithm based on FPGA, which is able to process large images at a fast speed. The algorithm avoids new dependencies by using a new sample ordering method and encoder.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Materials Science, Textiles
Feng Li, Qinggang Xi
Summary: This paper proposes a detection system suitable for textile printing positioning, which combines a neural network and FPGA to achieve rapid and accurate printing positioning. Experimental results show that the design scheme has lower power consumption compared to graphic processing units and faster speed compared to central processing units.
TEXTILE RESEARCH JOURNAL
(2022)
Article
Engineering, Mechanical
Zhicai Hu, Jiang Wang, Xinyu Hao, Kai Li
Summary: This study presents a hardware efficient, scalable, and real-time computing strategy for implementing large-scale motif cortical network via FPGA. The information transmission in cortical network is explored from the perspective of stochastic resonance, and it is found that the network can only transmit signal when the E/I balance is lopsided. The time delay between populations mainly affects information transmission by determining the phase timestamp of signal transmission.
NONLINEAR DYNAMICS
(2022)
Article
Engineering, Electrical & Electronic
Riccardo Della Sala, Davide Bellizia, Giuseppe Scotti
Summary: This paper presents a True Random Number Generator (TRNG) implemented using latched-XOR (LX) gates. The proposed TRNG improves the throughput of conventional TRNGs by combining latches metastability and ring oscillators jitter. Experimental results show that the generated bitstreams exhibit good randomness and the TRNG is robust to voltage and temperature variations. The FPGA implementation is compact and efficient, achieving high throughput.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2022)
Article
Chemistry, Analytical
Seda Guzel Aydin, Hasan Sakir Bilge
Summary: This study proposes an FPGA-based ultrasound image registration CNN (FUIR-CNN) to regress three rigid registration parameters from image pairs. The estimation process is accelerated using fixed-point data and parallel operations carried out by unrolling and pipelining techniques. Experimental results show that the FUIR-CNN achieves a 139 times faster inference phase compared to the software-based network, with a negligible drop in regression performance at a clock frequency of under 200 MHz. The proposed FPGA-based accelerated CNN offers high speed for registration parameters, less power consumption compared to CPU, and potential for real-time medical imaging.
Article
Computer Science, Information Systems
Miroslaw Chmiel, Robert Czerwinski, Andrzej Malcher
Summary: This article presents the designs of timer function blocks (FBs) and their implementation options, including timer-on, timer-off, and timer-pulse types. Both hardware and software-like designs are discussed, and they can function as multi-channel timers. The software-like design is particularly noteworthy as it eliminates the need for edge detectors. The timers were implemented using Verilog language on an FPGA chip, and the universal interface design allows them to be used for hardware support of existing PLCs or as integral parts of new PLC CPUs.
Article
Radiology, Nuclear Medicine & Medical Imaging
E. Dhiravidachelvi, Senthil S. Pandi, R. Prabavathi, Bala C. Subramanian
Summary: Diabetic retinopathy is a major cause of visual impairment in diabetes patients. Developing an automated decision-making system to predict the presence of exudates in fundus images can effectively improve prediction accuracy.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Engineering, Biomedical
R. Rajagopal, R. Karthick, P. Meenalochini, T. Kalaichelvi
Summary: A new method for lung disease detection is proposed in this paper, which improves the accuracy and performance by making improvements in pre-processing, feature extraction, and classifier optimization.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
J. Jasper Gnana Chandran, R. Karthick, R. Rajagopal, P. Meenalochini
Summary: This paper proposes a new method for bone age assessment called DCCGAN-GEO-BAA-HX-ray. The method improves accuracy and reduces computational time by using Tsallis entropy and Golden eagle optimization.
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
J. Karthika, A. Senthilselvi
Summary: In the real-world, e-commerce technologies allow people to easily select desired products and services. However, this technology also creates opportunities for scammers to commit credit card fraud. To prevent such fraudulent activities and payment losses, researchers have developed an automated system called CCFD for credit card fraud detection. This research work proposes a CNN-GRU model with a Navo Minority Over-Sampling Technique (NMOTe) to address the issue of class imbalance and achieve high accuracy in detecting credit card fraud.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Hardware & Architecture
R. Karthick, A. Senthilselvi, P. Meenalochini, S. Senthil Pandi
Summary: Partitioning and Floor Planning are two design processes used in VLSI design to reduce circuit size. Physical design automation aims to reduce area and interconnect length, thereby decreasing chip size. The proposed Optimal Partitioning and Floor Planning algorithm combines Whale Optimization and Adaptive Bird Swarm Optimization to achieve lower area, delay, and power usage compared to existing methods. Benchmark tests on MCNC circuits demonstrate the effectiveness of this hybrid algorithm.
JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS
(2023)
Article
Computer Science, Hardware & Architecture
P. Meenalochini, R. Karthick, E. Sakthivel
Summary: This paper proposes an effective hybrid control technique for an extended switched coupled inductor quasi-Z source inverter for 3F grid-connected photovoltaic system. The proposed hybrid system, named hybrid RERNN-CHGSO, combines Recalling Enhanced Recurrent Neural Network (RERNN) with Chaotic Henry Gas Solubility Optimization (CHGSO) to maximize power extraction and manage the performance of the PV system. The ESCL-quasi-Z-Source inverter modeling is improved to extract maximal power, and the proposed control system minimizes THD and injected power.
JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS
(2023)
Article
Computer Science, Information Systems
J. Karthika, A. Senthilselvi
Summary: Numerous organizations, including the financial industry, strongly support online service payments due to the rapid growth of internet commerce and banking. However, increasing levels of fraud and a decline in trust in online banking have led to significant global losses. Credit card fraud is a major concern, with illegal transactions being carried out by unauthorized users. The detection of credit card fraud is further challenged by the availability of public data, high false alarms, data imbalances, and evolving fraud patterns. Machine Learning techniques have been used for credit card fraud detection (CCFD) but have not provided satisfactory results. To address these issues, Deep Learning (DL) is now being applied to CCFD. This research work proposes a one-dimensional Dilated Convolutional Neural Network (DCNN) that learns both spatial and temporal features to improve the efficiency of CCFD, by implementing a dilated convolutional layer (DCL) and utilizing under-sampling and over-sampling techniques to address data imbalance. The proposed DCNN model with sampling techniques achieved an accuracy of 97.39% on a small card database, outperforming the existing CNN model, which achieved 94.44% accuracy on the same database.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Telecommunications
K. Ramya, Senthilselvi Ayothi
Summary: Cloud computing is a rapidly growing technology that provides virtualized computer resources to users through service providers. Load balancing and task scheduling are important concerns in cloud computing from the service provider's perspective. This article proposes a hybrid dingo and whale optimization algorithm-based load balancing mechanism (HDWOA-LBM) to improve resource utilization, reliability, and throughput in the cloud. The HDWOA-LBM mimics the hunting characteristics of dingo and VMs, utilizing the exploration and exploitation process to allocate incoming tasks to suitable VMs. Simulation experiments using CloudSim show that the proposed HDWOA-LBM achieves better throughput, reliability, makespan, and resource allocation compared to other intelligent load balancing schemes.
TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES
(2023)
Article
Engineering, Biomedical
B. Kalpana, A. K. Reshmy, S. Senthil Pandi, S. Dhanasekaran
Summary: Skin disease is the most common and dangerous disease, and the development of efficient and reliable skin cancer prediction techniques is necessary to prevent it. This paper proposes an ensemble support vector kernel random forest-based hybrid equilibrium Aquila optimization (ESVMKRF-HEAO) approach, and evaluates the model using the HAM10000 dataset. The experimental results show that the proposed model achieves high performance in accurately predicting and classifying skin lesion images.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Environmental Sciences
Senthil Pandi Sankareshwaran, Gitanjali Jayaraman, Pounambal Muthukumar, Arivuselvan Krishnan
Summary: Rice is a crucial cereal food crop for the majority of the world's population and its yield and quality are impacted by various biotic and abiotic factors. Rice plant disease is a significant concern in the agricultural sector, leading to losses in multiple aspects. To address this issue, a novel approach named CAHA-AXRNet is proposed, which optimizes the hyperparameters of the AX-RetinaNet model using the crossover boosted artificial hummingbird algorithm. The approach achieves an accuracy rate of 98.1% in rice plant disease detection.
ENVIRONMENTAL MONITORING AND ASSESSMENT
(2023)
Article
Engineering, Electrical & Electronic
G. Renith, A. Senthilselvi
Summary: Skin cancer, caused by abnormal and uncontrolled cell growth, is the most common deadly disease. Early identification is crucial, and this paper proposes an intelligent system to detect and classify dermoscopic images of skin lesions as malignant or benign. The proposed method preprocesses the images, extracts significant patterns using the AlexNet architecture, and utilizes the IAB-AAM classification model for discrimination.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2023)
Article
Multidisciplinary Sciences
V Surya, A. Senthilselvi
Summary: Natural oils such as avocado oil, corn oil, chamomile oil, and rapeseed oil have been used for centuries in different parts of the world for health, skin, and hair care. This paper presents a modified faster region-based convolutional neural network model for detecting oil adulteration, which has shown effectiveness in quickly identifying adulterated components in high-quality oils.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2023)