International Journal of Reconfigurable and Embedded Systems (IJRES)
Not a member yet
454 research outputs found
Sort by
Comparative analysis of ZigBee, LoRa, and NB-IoT in a smart building: advantages, limitations, and integration possibilities
Get PDFThis paper compares the performance of various wireless technologies: ZigBee, long range (LoRa), and narrowband internet of things (NB-IoT), which support smart building applications. The highlight of this work is that we focus on wireless communication between the floors of the building by analyzing the performance metrics using the received signal strength indicator (RSSI) and packet loss ratio (PLR). First, the ZigBee tests confirmed reliable packet delivery without any loss over distances up to 40 meters on the same floor, with RSSI results ranging from -65.5 to -87.5 dBm. ZigBee also maintained signal transmission through one cross-floor level, with RSSI values between -60 and -119 dBm. The second set of tests, with LoRa, indicated signal transmission over several floors with slightly improved RSSI values for the 2 dBi antenna compared to those for the -4 dBi antenna, despite increased packet loss with distance. Finally, NB-IoT showed the most consistent long-range connectivity, achieving a stable signal up to 458 meters from the base station with RSSI levels varying from -55.6 to -74.6 dBm, without packet loss in all tests. This study demonstrates how such technologies could be used in smart buildings and provides suggestions on how to determine the most suitable systems and configure them to ensure reliable communication networks within the building
Central processing unit load reduction through application code optimization and memory management
Get PDFCentral processing unit (CPU) loading refers to the amount of processing power a CPU uses to execute a given set of commands or perform an exact task. Higher CPU load can lead to slower, sluggish performance, reduced lifespan, and reduced system stability. Using the CPU Load trace results, the performance bottlenecks can be identified and suitable methods can be adopted to reduce the load on the CPU. For an ideal embedded system, the CPU should be in idle state for around 70% of CPU usage time. In this paper, three types of optimization techniques are implemented, which include application code optimization, memory management, and implementing interrupt-driven data transfer. Application code can be optimized by getting rid of redundant code, duplicate functions and function inlining, function cloning which reduces the size of the code with increase in reusability. By moving the data, variables to data tightly coupled memory (DTCM) and instructions, functions to instruction tightly coupled memory (ITCM), the speed of the CPU increases which reduces the load on CPU. The conventional polling method which increases the CPU load can be reduced by implementing the same in interrupt-driven data transfer. The load on the CPU has reduced from 89.53% to 29.58%
Design of agrivoltaic system with internet of things control for chili fruit classification using the neural network method
Get PDFAgriculture is a leading sector in the economy as well as the most dominant provider of employment for the Indonesian people. The fertile soil factor allows various types of fruit to be grown, including chilies. However, complex problems make chili farmers have limitations in implementing conventional farming systems. Therefore, the development of an agrivoltaic system with internet of things (IoT) integrated sensors on chili plants can help farmers more easily control, add vitamins, fertilizers, and provide plant nutrients that can be done automatically periodically based on a real-time clock schedule. This system also operates using photovoltaic (PV) as a pumping machine for water circulation. Other technologies such as mini smart cameras are also being developed to monitor and take pictures of chilies which will later be converted using the graphical user interface (GUI) application for segmentation. The method used in this chili fruit classification uses an artificial neural network in classifying ripe, raw, and rotten chilies. The classification results obtained an R value of 0.9, which means it is close to a value of 1 in the suitability of the chili image. Therefore, farmers will find it easier to sort the chilies that will be harvested
Multimodal recognition with deep learning: audio, image, and text
Get PDFEmotion detection is essential in many domains including affective computing, psychological assessment, and human computer interaction (HCI). It contrasts the study of emotion detection across text, image, and speech modalities to evaluate state-of-the-art approaches in each area and identify their benefits and shortcomings. We looked at present methods, datasets, and evaluation criteria by conducting a comprehensive literature review. In order to conduct our study, we collect data, clean it up, identify its characteristics and then use deep learning (DL) models. In our experiments we performed text-based emotion identification using long short-term memory (LSTM), term frequency-inverse document frequency (TF-IDF) vectorizer, and image-based emotion recognition using a convolutional neural network (CNN) algorithm. Contributing to the body of knowledge in emotion recognition, our study's results provide light on the inner workings of different modalities. Experimental findings validate the efficacy of the proposed method while also highlighting areas for improvement
Design and implementation of smart traffic light controller with emergency vehicle detection on FPGA
Get PDFIncreased traffic volumes resulting from urbanization, industrialization, and population growth have given rise to complex issues, including congestion, accidents, and traffic violations at intersections. In the absence of a functional smart traffic light system, traffic congestion occurs due to imbalanced traffic flow at intersections. Current traffic management lacks provisions for ensuring the unobstructed movement of emergency vehicles, even a small delay for which can have significant consequences. This paper presents a smart traffic light controller developed using Verilog hardware description language (HDL) in Quartus Prime 21.1 and Questa Intel field programmable gate array (FPGA) Starter Edition 2021.2, and implemented on an Altera DE2-115 FPGA. The controller is designed specifically to detect emergency vehicle at four-way intersections for inputs radio frequency identification (RFID) readers and infrared (IR) sensors. The RFID readers and IR sensors are managed through slide switches on the FPGA board. The smart traffic light controller contains three sub-modules: clock division, counter, and finite state machine (FSM) operation, enabling it to manage traffic in scenarios with emergency vehicles, high traffic density, and low traffic density. This proposed system can alleviate intersection congestion by controlling access and allocating time effectively. In conclusion, the project ensures the smooth passage of emergency vehicles by continuously monitoring their presence and giving them priority in traffic flow
Enhancing scalability and efficiency in technological transaction utilizing dual-layer blockchain approach
Get PDFThe leather industry encounters significant challenges in integrating blockchain technology and smart contracts into its complex supply networks. Despite technological advancements, existing supply chain management systems suffer from inefficiencies, opacity, and vulnerabilities to fraud. Blockchain offers promising solutions such as immutable ledgers, decentralized governance, and smart contract automation. However, scalability limitations hinder the efficient handling of high transaction volumes, impacting procurement, production, inventory management, and distribution processes, leading to delays and increased costs. This research aims to address these challenges by exploring innovative approaches, including dual-layer blockchain architectures incorporating sharding and state channels, tailored to the unique needs of the leather industry. By overcoming scalability barriers, the research seeks to unlock the transformative potential of blockchain technology and smart contracts, enhancing transparency, traceability, and efficiency in leather supply chains while ensuring global interoperability and regulatory compliance. Through empirical validation and comparative analysis, this study provides understandings into the practical implementation of blockchain solutions within the leather industry, offering strategic guidance for sustainable supply chain management practices
Investigating the performance of RNN model to forecast the electricity power consumption in Guangzhou China
Get PDFThe project initiatives to create a reliable prediction model for power loads in Guangzhou, China. The power industry is facing issues due to rapid market growth and the necessity for better grid management, prompting this response. In developing the models, conventional machine learning models have been used so far, but in this study, the performance of deep learning is investigated. Therefore, the recurrent neural network (RNN) was selected for the prediction of electricity consumption. Later, the performance of the model was compared with autoregressive integrated moving average (ARIMA), long short-term memory (LSTM), and RNN. The experimental results show that the RNN outperforms ARIMA and LSTM, with an R² value of 0.92, an RMSE of 0.13107 and an MAE of 0.0176. The project improved power resource planning and management, selected an acceptable forecasting model RNN and contributed to forecasting technology developments. The study identified limits in historical data availability and quality, as well as external influences affecting the studies. RNN models can help optimize resource allocation and improve energy planning
Reconfigurable embedded systems for remote health monitoring: a comprehensive review
Get PDFThe rapid expansion of telemedicine and wearable health devices has intensified the demand for energy-efficient and adaptable embedded systems capable of supporting real-time, reliable remote health monitoring. This review provides a comprehensive survey of reconfigurable embedded platforms—focusing on field-programmable gate arrays (FPGAs), coarse-grained reconfigurable arrays (CGRAs), and heterogeneous system-on-chips (SoCs)—deployed for monitoring critical physiological parameters such as electrocardiogram (ECG), oxygen saturation (SpO₂), and body temperature. We analyze co-design methodologies that integrate artificial intelligence (AI-driven) neural accelerators, quantization strategies, and runtime adaptability to address the competing requirements of low power consumption, data integrity, and latency minimization in diverse telemedicine contexts. The paper highlights the strengths and limitations of conventional versus reconfigurable approaches, reviews case studies in wearable and implantable health devices, and underscores key design trade-offs in performance, scalability, and security. By systematically mapping current innovations and identifying unresolved challenges—including standardization, clinical validation, and secure edge integration—this review positions reconfigurable architectures as a cornerstone for next-generation, patient-centric remote health monitoring. Future directions emphasize AI-enabled adaptability, sustainable and carbon-aware device design, and personalized healthcare through adaptive embedded systems, charting a pathway toward scalable and resilient telemedicine ecosystems
Enhancing cross-cutting concerns in the internet of things with applying aspect oriented programming
Get PDFAspect oriented programming (AOP) is a new programming model that provides new concepts to handle cross-cutting concerns about code. The idea of introducing AOP in the internet of things (IoT) is inherited from the complexity of sensor operations involving data acquisition, processing, and communication, the need to support multiple simultaneous services for users particularly security services such as authentication, authorization, data traceability, and transaction management, and the challenges posed by the IoT deployments, the treatment of these data volumes lead to problematic code redundancy and cross-cutting concerns that compromise system maintainability. In this context, AOP enables the separation of core functionalities, data management, and cross-cutting concerns, allowing them to be developed and reused independently within the same codebase. To address these issues, this paper proposes an AOP model for IoT systems based on the Petri net representations. The model strategically integrates the core AOP advantages of modularity, reusability, and extensibility, microservices based architectural decomposition and specialized handling of sensor-specific requirements in IoT environments
Real-time face detection and local binary patterns histograms-based face recognition on Raspberry Pi with OpenCV
Get PDFThis paper presents a practical end-to-end paper demonstrating real-time face recognition using a Raspberry Pi and open source computer vision library (OpenCV) consisting of three main stages: training the recognizer, real-time recognition, and face detection and data gathering. The paper offers a comprehensive guide for enthusiasts venturing into computer vision and facial recognition. Employing the Haar Cascade classifier for accurate face detection and the local binary patterns histograms (LBPH) face recognizer for robust training and recognition, the paper ensures a thorough understanding of key concepts. The step-by-step process covers software installation, camera testing, face detection, data collection, training, and real time recognition. With a focus on the Raspberry Pi platform, this paper serves as an accessible entry point for exploring facial recognition technology. Readers will gain insights into practical implementation, making it an ideal resource for learners and hobbyists interested in delving into the exciting realm of computer vision