International Journal of Reconfigurable and Embedded Systems (IJRES)
Not a member yet
454 research outputs found
Sort by
Parallel graph algorithms on a RISCV-based many-core
Get PDFGraph algorithms are essential in domains like social network analysis, web search, and bioinformatics. Their execution on modern hardware is vital due to the growing size and complexity of graphs. Traditional multi-core systems struggle with irregular memory access patterns in graph workloads. Reduced instruction set computer–five (RISC-V)-based many-core processors offer a promising alternative with their customizable open-source architecture suitable for optimization. This work focuses on parallelizing graph algorithms like breadth-first search (BFS) and PageRank (PR) on RISC-V many-core systems. We evaluated performance based on graph structure and processor architecture, and developed an analytical model to predict execution time. The model incorporates the unique characteristics of the RISC-V architecture and the types and numbers of instructions executed by multiple cores, with a maximum prediction error of 11%. Our experiments show a speedup of up to 11.55× for BFS and 7.56× for PR using 16 and 8 cores, respectively, over single-core performance. Comparisons with existing graph processing frameworks demonstrate that RISC-V systems can deliver up to 20× better energy efficiency on real-world graphs from the network repository
Design and optimization of bail-shaped microstrip patch antenna for mid-band 5G application using a lightGBM model
Get PDFThis study suggests a bail-shaped microstrip patch antenna designed for 5G applications. This antenna model operates in the 3.45 GHz wireless communication frequency range, which is a component of the so-called C-band (3.3 to 4.2 GHz), which is widely utilized for mid-band 5G deployments across the globe. Antenna size optimization is achieved at 31×28 mm2. On the patch, a slot is added to enhance the return loss features. The light gradient boosting machine (LightGBM) model for prediction acts as an objective function of the considered piranha foraging optimization algorithm (PFOA) to adjust the antenna's slot dimension, which will be used to optimize the slot width. In order to get a superior return loss value of around -39.90<-10 dB, the optimization approach that is provided seeks to achieve the ideal slot length. The proposed device exhibits remarkable radiation efficiency by partially grounding, with a peak gain of around 2.535 dBi at 3.45 GHz. A novel hybrid approach combines the LightGBM prediction model with the PFOA to fine-tune slot dimensions, achieving a superior return loss of -39.90 dB. The exclusivity of this effort is the incorporation of machine learning algorithms to attain significantly improved parameters
FPGA implementation of artificial neural network for PUF modeling
Get PDFField-programmable gate array (FPGA) is a prominent device in developing the internet of things (IoT) application since it offers parallel computation, power efficiency, and scalability. The identification and authentication of these FPGAbased IoT applications are crucial to secure the user-sensitive data transmitted over IoT networks. Physical unclonable function (PUF) technology provides a great capability to be used as device identification and authentication for FPGAbased IoT applications. Nevertheless, conventional PUF-based authentication suffers a huge overhead in storing the challenge-response pairs (CRPs) in the verifier’s database. Therefore, in this paper, the FPGA implementation of the Arbiter-PUF model using an artificial neural network (ANN) is presented. The PUF model can generate the CRPs on-the-fly upon the authentication request (i.e., by a prover) to the verifier and eliminates huge storage of CRPs database in the verifier. The architecture of ANN (i.e., Arbiter-PUF model) is designed in Xilinx system generator and subsequently converted into intellectual property (IP). Further, the IP is programmed in Xilinx Artix-7 FPGA with other peripherals for CRPs generation and validation. The findings show that the Arbiter-PUF model implementation on FPGA using the ANN technique achieves approximately 98% accuracy. The model consumes 12,196 look-up tables (LUTs) and 67 mW power in FPGA
Implementation of hardware security module using elliptic curve cryptography for cyber-physical system
Get PDFThe vision of sustainable development goal 9 (SDG 9) is realized through the integration of innovative technologies in the cyber-physical system (CPS). This work focuses on a smart network meter (SNM) application, designed to manage the extensive big data analytics required for processing and analyzing vast amounts of aggregated data in a short period. To address these demands, an advanced explicitly parallel instruction computing (AEPIC) approach is employed, leveraging a multi-core hardware security module (HSM) built on the elliptic curve cryptography (ECC) algorithm. Implementing the algorithm on various field programmable gate arrays (FPGAs) ensures adaptability to different hardware configurations, delivering scalable and optimized performance for big data aggregation in SNM applications. The proposed module showcases exceptional performance in design analysis. The Virtex-7 FPGA demonstrates excellent suitability for big data analytics in smart network applications, with dynamic power consumption accounting for 55% of total power and an on-chip power of 0.542 watts
A k-nearest neighbors algorithm for enhanced clustering in wireless sensor network protocols
Get PDFWireless sensor networks (WSNs) are small, autonomous, battery-powered nodes capable of sensing, storing, and processing data, while communicating wirelessly with a central base station (BS). Optimizing energy consumption is a major challenge to extend the lifetime of these networks. In this study, we propose an innovative approach combining the k-nearest neighbors (KNN) algorithm with hierarchical and flat routing protocols to improve node selection and clustering in three key protocols: low-energy adaptive clustering hierarchy (LEACH), threshold-sensitive energy efficient sensor network protocol (TEEN), and hybrid energy-efficient distributed clustering (HEED). Concretely, KNN is used to rank nodes based on their spatial and energy proximity, thus optimizing the choice of cluster heads (CHs) and reducing long and costly connections. Simulations show a reduction in the inter-CH distance, a decrease in overall energy consumption, and an extension of the network lifetime compared to conventional versions of the protocols. These improvements not only help increase operational efficiency, but also enhance communications stability and security, providing a robust and sustainable solution for critical WSN applications
The novel single-module communication subsystem architecture for industrial digital inkjet
Get PDFThe typical challenge in embedded hardware development is the data transfer subsystem. As long as the required speeds are low and high latency is acceptable, there is quite a simple solution with serial bus like controller area network (CAN). In case of high speed (hundreds of megabits per second) with the high temporal determinism, the solution becomes significantly more complicated, requiring expensive components and growing complexity of the embedded software/firmware. We consider industrial inkjet as an example. The device typically includes moving carriage (with printheads) to jet along the media. Existing solutions use optical fiber cable or shielded twisted pair (STP) cable to connect modules. So, additional physical and logical devices are required (for example, for buffering or serial-to-parallel data conversion). For a long time, this approach has no valuable alternative. The novel single-module solution involves abandoning the intermediate high-speed channel. Instead of multiple modules and high-speed communication links between them, the single module is installed near the data destination and connected to the master PC via Ethernet. The functionality of high-speed data transfer subsystem is delegated to the shared dynamic random-access memory (DRAM) and controller, implemented with field-programmable gate array (FPGA) resources. So, the connection cable is not needed anymore and the transfer speed is virtually limited only by DRAM performance
Development of a web-based application for real-time eye disease classification system using artificial intelligence
Get PDFThe incorporation of artificial intelligence (AI) into the field of medicine has created new strategies in enhancing the detection of disease, with a focus on the identification of eye diseases such as glaucoma, diabetic retinopathy, and macular degeneration associated with age, which can lead to blindness if not detected and treated early enough. Driven by the need to combat blindness, which affects approximately 39 million people globally, according to the World Health Organization (WHO). This research offers a web-based, real time approach to classifying eye diseases from fundus images due to user friendliness. Three pre-trained convolutional neural network (CNN) models are adopted, namely ResNet-50, Inception-v3, and MobileNetV3. The models were trained on a dataset of 8000 fundus images subdivided into four classes: cataract, glaucoma, diabetic retinopathy, and normal eyes. The performance of the models was evaluated in 3-way (normal eye and two diseases) and 4-way (normal eye and three diseases). ResNet-50 had higher performances, with 98% and 97% accuracy in the respective classifications, compared to InceptionV3 and MobileNetV3. Consequently, ResNet-50 was used in an online application that made real-time diagnoses. This research findings reveal the potential of CNNs in the healthcare industry, particularly in reducing over-reliance on specialists and increasing access to quality diagnostic technologies. Especially in critical areas such as this with limited healthcare resources, where the technology can create significant gaps in disease detection and control
Waste incinerator monitoring system based on remote communication with android interface
Get PDFRaya Ngijo Housing, one of the areas in Karangploso in Malang District has a temporary waste management team that organises the collection of waste from residents and sends it to the landfill. The process of collecting waste from residents is usually at the temporary disposal site (TPS) in the form of moving waste from residential cleaning vehicles and accommodated at the TPS until collection by the Malang District environmental service container for disposal to the transferred to landfills (TPA). Problems often occur when the container collection process is delayed for various reasons, so that the amount of rubbish in the TPS is excessive. One of the solutions made by the cleaning team is to burn excess waste and can be burned using a furnace. However, the combustion carried out cannot be ensured perfect combustion which is feared by the environmental service. Therefore, a remote communication-based furnace monitoring system and android application were made to ensure the perfection of the combustion process so that it could be monitored by the cleaning team. Parts per million (PPM) carbon dioxide (CO2) levels of combustion smoke and combustion temperature are also monitored and controlled in accordance with the safe standards set by the environmental agency
Video surveillance system based on artificial vision and fog computing for the detection of lethal weapons
Get PDFCitizen insecurity in underdeveloped third world countries is aggravated by poor management of arms control and illegal trafficking, which requires information technology solutions in intelligent video surveillance systems for the detection of lethal weapons. The literature review highlights the need for an intelligent video surveillance system to combat high crime, using fog computing, which processes data in real time for the detection of weapons and other crimes. Furthermore, at an international level, solutions based on artificial intelligence and deep learning are being implemented for object recognition and weapons detection. Therefore, this paper describes the design of an intelligent video surveillance system based on artificial vision, fog and edge computing to detect lethal weapons in domestic environments, performing weapon classification and data transmission to police centers. The intelligent video surveillance system allows detecting lethal weapons and operates in three stages: an edge node with a Raspberry Pi 4; a detection algorithm based on a convolutional neural network with YOLOv5; and streaming tagged images to a security unit via WhatsApp. The main conclusion is that the system achieved a precision greater than 0.85 and a quick and efficient response in sending alerts, representing a scalable and effective solution against home burglary
Comparative analysis of feature descriptors and classifiers for real-time object detection
Get PDFDetecting objects within complex environments, such as urban settings, holds significant importance across various applications, including driver assistance systems, traffic monitoring, and obstacle detection systems. Particularly crucial for these applications is the accurate differentiation between cars and roads. This study introduces a novel approach that leverages traditional feature descriptors and classifiers for real-time object detection. It conducts an exhaustive comparative analysis of feature descriptors and classifiers to identify the most effective model for real-time object detection. Handcrafted features of images are extracted using algorithms such as scale invariant feature transform (SIFT), oriented fast and brief (ORB), fast retina key-point (FREAK), and local binary pattern (LBP). Seven classifiers are employed, including support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), decision tree (DT), logistic regression (LR), Naive Bayes, and extreme gradient boosting (XGBoost). The performance of the 28 generated combinations of feature descriptors and classifiers is evaluated based on the parameters of accuracy, precision, F1 score, and recall. The model utilizing LBP and XGBoost achieves the highest accuracy, reaching 83.59%. The system architecture comprises a camera, a high-speed computing unit, a display, and an audio subsystem, with the algorithm implemented on a Raspberry Pi 4B (8 GB)