Journal of Sensor and Actuator Networks

30 pages, 6468 KiB

Open AccessArticle

EWOD Sensor for Rapid Quantification of Marine Dispersants in Oil Spill Management

by Oriol Caro-Pérez, María Blanca Roncero and Jasmina Casals-Terré

J. Sens. Actuator Netw. 2025, 14(3), 54; https://doi.org/10.3390/jsan14030054 - 21 May 2025

In this study, we introduce a novel Electrowetting-on-Dielectric (EWOD) sensor designed to quantify marine dispersants at the spill point. The sensor quantifies changes in the surface tension of liquid droplets at varying dispersant concentrations through the deformation response of the droplet under applied [...] Read more.

In this study, we introduce a novel Electrowetting-on-Dielectric (EWOD) sensor designed to quantify marine dispersants at the spill point. The sensor quantifies changes in the surface tension of liquid droplets at varying dispersant concentrations through the deformation response of the droplet under applied voltage. Analyzed responses include droplet length and contact angle (CA) on the device surface upon sensor activation. This sensor offers significant advantages over existing chemical methods, which are costly and complex. Moreover, compared to conventional methods based on the same principle, it demonstrates enhanced sensitivity at low concentrations. Additionally, the sensor’s portability enables instantaneous and in situ measurements of marine dispersant concentrations, thus providing a crucial tool for effective oil spill response by facilitating on-site decision-making and offering higher temporal resolution for studies on the marine dispersant’s environmental impact. The device’s potential extends beyond marine dispersants to detecting various contaminants affecting surface tension. Its adaptability underscores the EWOD device’s role as a versatile tool for environmental monitoring and on-site analysis, addressing the urgent need for efficient and sustainable solutions in environmental management. Full article

► Show Figures

Figure 1

26 pages, 2634 KiB

Open AccessArticle

Optimized Dual-Battery System with Intelligent Auto-Switching for Reliable Soil Nutrient Monitoring in Remote IoT Applications

by Doan Perdana, Pascal Lorenz and Bagus Aditya

J. Sens. Actuator Netw. 2025, 14(3), 53; https://doi.org/10.3390/jsan14030053 - 19 May 2025

Abstract

This study introduces a novel dual-battery architecture with intelligent auto-switching control, designed to ensure uninterrupted operation of agricultural sensing systems in environments with unpredictable energy availability. The proposed system integrates Lithium-Sulphur (Li-S) and Lithium-Ion (Li-Ion) batteries with advanced switching algorithms—specifically, the Dynamic Load [...] Read more.

This study introduces a novel dual-battery architecture with intelligent auto-switching control, designed to ensure uninterrupted operation of agricultural sensing systems in environments with unpredictable energy availability. The proposed system integrates Lithium-Sulphur (Li-S) and Lithium-Ion (Li-Ion) batteries with advanced switching algorithms—specifically, the Dynamic Load Balancing–Power Allocation Optimisation (DLB–PAO) and Dynamic Load Balancing–Genetic Algorithm (DLB–GA)—tailored to maximise sensor operational longevity. By optimizing the dual-battery configuration for real-world deployment and conducting comparative evaluations across multiple system designs, this work advances an innovative engineering solution with significant practical implications for sustainable agriculture and remote sensing applications. Unlike conventional single-battery systems or passive redundancy approaches, the architecture introduces active redundancy, adaptive energy management, and fault tolerance, substantially improving operational continuity. A functional prototype was experimentally validated using realistic load profiles, demonstrating seamless battery switching, extended uptime, and enhanced energy reliability. To further assess long-term performance under continuous Internet of Things (IoT) operation, a simulation framework was developed in MATLAB/Simulink, incorporating battery degradation models and empirical sensor load profiles. The experimental results reveal distinct performance improvements. A baseline single-battery system sustains 28 h of operation with 31.2% average reliability, while a conventional dual-battery configuration extends operation to 45 h with 42.6% reliability. Implementing the DLB–PAO algorithm elevates the average reliability to 91.7% over 120 h, whereas the DLB–GA algorithm achieves near-perfect reliability (99.9%) for over 170 h, exhibiting minimal variability (standard deviation: 0.9%). The integration of intelligent auto-switching mechanisms and metaheuristic optimisation algorithms demonstrates a marked enhancement in both reliability and energy efficiency for soil nutrient monitoring systems. This method extends the lifespan of electronic devices while ensuring reliable energy storage over time. It creates a practical foundation for sustainable IoT agricultural systems in areas with limited resources. Full article

► Show Figures

Figure 1

34 pages, 792 KiB

Open AccessArticle

Data-Driven Approaches for Efficient Vehicle Driving Analysis: A Survey

by Iryna I. Husyeva, Ismael Navas-Delgado and José García-Nieto

J. Sens. Actuator Netw. 2025, 14(3), 52; https://doi.org/10.3390/jsan14030052 - 19 May 2025

Abstract

Efficient vehicle driving generally intends to reduce fuel consumption, emissions of harmful substances, and accident rates based on energy-efficient driving patterns as a set of parameters defining optimal vehicle and route characteristics, together with specific ways of driving a vehicle that the particular [...] Read more.

Efficient vehicle driving generally intends to reduce fuel consumption, emissions of harmful substances, and accident rates based on energy-efficient driving patterns as a set of parameters defining optimal vehicle and route characteristics, together with specific ways of driving a vehicle that the particular driver applies. To gain environmental friendliness in driving, two main approaches can be outlined: optimal route planning and driver training based on the principles of ecological driving. The latter can be supported by using software for real-time, efficient vehicle driving recommendations. In order to develop the principles of ecological driving as well as generate relevant real-time recommendations, it is necessary to identify the specific parameters required to analyze driver behavior and vehicle performance, determine the corresponding energy consumption, and understand the influence of route and environmental conditions on overall efficient vehicle driving. These tasks require a large amount of data, often obtained from heterogeneous sources, which, when publicly available, are complex for consolidation, transmission, and processing, not to mention the complexity of the data model itself. This study provides a thorough review of the current data sources and techniques for efficient vehicle driving analysis, focusing on the availability and relevance of dataset sources and repositories. The categorization of parameters and data processing techniques enabling efficient vehicle driving analysis is carried out according to efficiency types such as driver’s efficiency, resource consumption efficiency, and route planning efficiency. For each type of efficiency, we provide a list of contextual groups and features, identifying the dataset containing the necessary feature, making it possible not only to determine the parameters defining, for example, driver efficiency, but also locate the corresponding dataset serving as a stepping stone for researchers and practitioners to join the community investigating efficient vehicle driving analysis. We also discuss future trends and perspectives, identifying alternative data sources for efficient vehicle driving analysis, and focus on data collection issues revealed by the practical use case of collecting data from mobile phone sensors. Full article

(This article belongs to the Special Issue Advances in Intelligent Transportation Systems (ITS))

► Show Figures

Figure 1

20 pages, 7183 KiB

Open AccessArticle

A Two-Stage Strategy Integrating Gaussian Processes and TD3 for Leader–Follower Coordination in Multi-Agent Systems

by Xicheng Zhang, Bingchun Jiang, Fuqin Deng and Min Zhao

J. Sens. Actuator Netw. 2025, 14(3), 51; https://doi.org/10.3390/jsan14030051 - 14 May 2025

Abstract

In mobile multi-agent systems (MASs), achieving effective leader–follower coordination under unknown dynamics poses significant challenges. This study proposes a two-stage cooperative strategy that integrates Gaussian Processes (GPs) for modeling and a Twin Delayed Deep Deterministic Policy Gradient (TD3) for policy optimization (GPTD3), aiming [...] Read more.

In mobile multi-agent systems (MASs), achieving effective leader–follower coordination under unknown dynamics poses significant challenges. This study proposes a two-stage cooperative strategy that integrates Gaussian Processes (GPs) for modeling and a Twin Delayed Deep Deterministic Policy Gradient (TD3) for policy optimization (GPTD3), aiming to enhance adaptability and multi-objective optimization. Initially, GPs are utilized to model the uncertain dynamics of agents based on sensor data, providing a stable and noiseless training virtual environment for the first phase of TD3 strategy network training. Subsequently, a TD3-based compensation learning mechanism is introduced to reduce consensus errors among multiple agents by incorporating the position state of other agents. Additionally, the approach employs an enhanced dual-layer reward mechanism tailored to different stages of learning, ensuring robustness and improved convergence speed. Experimental results using a differential drive robot simulation demonstrate the superiority of this method over traditional controllers. The integration of the TD3 compensation network further improves the cooperative reward among agents. Full article

► Show Figures

Figure 1

29 pages, 2615 KiB

Open AccessReview

A Review: Applications of MOX Sensors from Air Quality Monitoring to Biomedical Diagnosis and Agro-Food Quality Control

by Elisabetta Poeta, Estefanía Núñez-Carmona and Veronica Sberveglieri

J. Sens. Actuator Netw. 2025, 14(3), 50; https://doi.org/10.3390/jsan14030050 - 9 May 2025

Abstract

Metal oxide semiconductor (MOX) sensors are emerging as a groundbreaking technology due to their remarkable features: high sensitivity, rapid response time, low cost, and potential for miniaturization. Their ability to detect volatile organic compounds (VOCs) in real time makes them ideal tools for [...] Read more.

Metal oxide semiconductor (MOX) sensors are emerging as a groundbreaking technology due to their remarkable features: high sensitivity, rapid response time, low cost, and potential for miniaturization. Their ability to detect volatile organic compounds (VOCs) in real time makes them ideal tools for applications across various fields, including environmental monitoring, medicine, and the food industry. This paper explores the evolution and growing utilization of MOX sensors, with a particular focus on atmospheric pollution monitoring, non-invasive disease diagnostics through the analysis of volatile compounds emitted by the human body, and food quality assessment. The crucial role of MOX sensors in monitoring the freshness of food and water, detecting chemical and biological contamination, and identifying food fraud is specifically examined. The rapid advancement of this technology offers new opportunities to improve quality of life, food safety, and public health, positioning MOX sensors as a key tool to address future challenges in these vital sectors. Full article

(This article belongs to the Section Actuators, Sensors and Devices)

► Show Figures

Figure 1

33 pages, 1438 KiB

Open AccessArticle

Mental Disorder Assessment in IoT-Enabled WBAN Systems with Dimensionality Reduction and Deep Learning

by Damilola Olatinwo, Adnan Abu-Mahfouz and Hermanus Myburgh

J. Sens. Actuator Netw. 2025, 14(3), 49; https://doi.org/10.3390/jsan14030049 - 7 May 2025

Abstract

Mental health is an important aspect of an individual’s overall well-being. Positive mental health is correlated with enhanced cognitive function, emotional regulation, and motivation, which, in turn, foster increased productivity and personal growth. Accurate and interpretable predictions of mental disorders are crucial for [...] Read more.

Mental health is an important aspect of an individual’s overall well-being. Positive mental health is correlated with enhanced cognitive function, emotional regulation, and motivation, which, in turn, foster increased productivity and personal growth. Accurate and interpretable predictions of mental disorders are crucial for effective intervention. This study develops a hybrid deep learning model, integrating CNN and BiLSTM applied to EEG data, to address this need. To conduct a comprehensive analysis of mental disorders, we propose a two-tiered classification strategy. The first tier classifies the main disorder categories, while the second tier classifies the specific disorders within each main disorder category to provide detailed insights into classifying mental disorder. The methodology incorporates techniques to handle missing data (kNN imputation), class imbalance (SMOTE), and high dimensionality (PCA). To enhance clinical trust and understanding, the model’s predictions are explained using local interpretable model-agnostic explanations (LIME). Baseline methods and the proposed CNN–BiLSTM model were implemented and evaluated at both classification tiers using PSD and FC features. On unseen test data, our proposed model demonstrated a 3–9% improvement in prediction accuracy for main disorders and a 4–6% improvement for specific disorders, compared to existing methods. This approach offers the potential for more reliable and explainable diagnostic tools for mental disorder prediction. Full article

(This article belongs to the Special Issue Applications of Wireless Sensor Networks: Innovations and Future Trends)

► Show Figures

Figure 1

22 pages, 2565 KiB

Open AccessReview

Exploring Robotic Technologies for Upper Limb Rehabilitation: Current Status and Future Directions

by Fabian Horacio Diaz, Carlos Borrás Pinilla and Cecilia E. García Cena

J. Sens. Actuator Netw. 2025, 14(3), 48; https://doi.org/10.3390/jsan14030048 - 1 May 2025

Abstract

This paper explores the design, control, construction, and leading manufacturers of upper limb rehabilitation robots through a thorough literature review. Utilizing databases such as Scopus, IEEE Xplore, Science Direct, Springer Link, and the Clinical Trials database, the research adhered to a rigorous screening [...] Read more.

This paper explores the design, control, construction, and leading manufacturers of upper limb rehabilitation robots through a thorough literature review. Utilizing databases such as Scopus, IEEE Xplore, Science Direct, Springer Link, and the Clinical Trials database, the research adhered to a rigorous screening process in accordance with PRISMA guidelines. This included analyzing abstracts and conducting comprehensive reviews of full articles when necessary. A total of fourteen relevant papers were systematically selected for in-depth analysis. The study offers a detailed classification of robotic technologies along with their Technology Readiness Levels (TRLs), discusses the primary challenges hindering their adoption, and proposes strategic research directions to address these issues. In conclusion, while upper limb robotic devices exhibit significant potential, persistent technological and design challenges must be addressed, underscoring the need for ongoing research and multidisciplinary collaboration to facilitate broader and more effective adoption. Full article

► Show Figures

Figure 1

23 pages, 3481 KiB

Open AccessArticle

Evaluating QoS in Dynamic Virtual Machine Migration: A Multi-Class Queuing Model for Edge-Cloud Systems

by Anna Kushchazli, Kseniia Leonteva, Irina Kochetkova and Abdukodir Khakimov

J. Sens. Actuator Netw. 2025, 14(3), 47; https://doi.org/10.3390/jsan14030047 - 25 Apr 2025

Abstract

The efficient migration of virtual machines (VMs) is critical for optimizing resource management, ensuring service continuity, and enhancing resiliency in cloud and edge computing environments, particularly as 6G networks demand higher reliability and lower latency. This study addresses the challenges of dynamically balancing [...] Read more.

The efficient migration of virtual machines (VMs) is critical for optimizing resource management, ensuring service continuity, and enhancing resiliency in cloud and edge computing environments, particularly as 6G networks demand higher reliability and lower latency. This study addresses the challenges of dynamically balancing server loads while minimizing downtime and migration costs under stochastic task arrivals and variable processing times. We propose a queuing theory-based model employing continuous-time Markov chains (CTMCs) to capture the interplay between VM migration decisions, server resource constraints, and task processing dynamics. The model incorporates two migration policies—one minimizing projected post-migration server utilization and another prioritizing current utilization—to evaluate their impact on system performance. The numerical results show that the blocking probability for the first VM for Policy 1 is 2.1% times lower than for Policy 2 and the same metric for the second VM is 4.7%. The average server’s resource utilization increased up to 11.96%. The framework’s adaptability to diverse server–VM configurations and stochastic demands demonstrates its applicability to real-world cloud systems. These results highlight predictive resource allocation’s role in dynamic environments. Furthermore, the study lays the groundwork for extending this framework to multi-access edge computing (MEC) environments, which are integral to 6G networks. Full article

(This article belongs to the Section Communications and Networking)

► Show Figures

Figure 1

41 pages, 5573 KiB

Open AccessReview

Socio-Organisational Challenges and Impacts of IoT: A Review in Healthcare and Banking

by Tahera Kalsoom, Naeem Ramzan, Shehzad Ahmed, Nadeem Anjum, Ghazanfar Ali Safdar and Masood Ur Rehman

J. Sens. Actuator Netw. 2025, 14(3), 46; https://doi.org/10.3390/jsan14030046 - 24 Apr 2025

Abstract

The Internet of Things (IoT) is transforming how organisations and individuals connect and interact with digital ecosystems, especially in sectors like healthcare and banking. While technological benefits have been widely discussed, the societal and organisational impacts of IoT adoption remain underexplored. This study [...] Read more.

The Internet of Things (IoT) is transforming how organisations and individuals connect and interact with digital ecosystems, especially in sectors like healthcare and banking. While technological benefits have been widely discussed, the societal and organisational impacts of IoT adoption remain underexplored. This study aims to address this gap by conducting a systematic literature review (SLR) of 110 peer-reviewed publications from 2012 to 2024 across four major academic databases. The review identifies and categorises the key applications of IoT, its social and organisational drivers, and the challenges of its implementation within the healthcare and banking sectors. The analysis reveals that critical barriers to IoT adoption include security, privacy, interoperability, and legal compliance, alongside concerns around workforce displacement and trust. This study also introduces the 5Cs framework—connectivity, continuity, compliance, coexistence, and cybersecurity—as a practical lens for addressing these challenges. The findings highlight the need for responsible IoT integration that balances innovation with ethical, social, and organisational accountability. Implications of this research inform policymakers, practitioners, and researchers on how to design human-centric and socially sustainable IoT strategies in sensitive sectors. Full article

► Show Figures

Figure 1

22 pages, 7750 KiB

Open AccessArticle

Coherence Analysis for Vibration Monitoring Under High Variability Conditions: Constraints for Cultural Heritage Preventive Conservation

by Claudia Pirrotta, Anna Maria Gueli, Carlo Trigona and Sebastiano Imposa

J. Sens. Actuator Netw. 2025, 14(2), 45; https://doi.org/10.3390/jsan14020045 - 21 Apr 2025

Abstract

The development of reliable sensor networks for vibration monitoring is essential for the preventive conservation of buildings and structures. The identification of natural frequencies is crucial both for sensor network planning, to ensure optimal placement, and for operation, to detect frequency shifts that [...] Read more.

The development of reliable sensor networks for vibration monitoring is essential for the preventive conservation of buildings and structures. The identification of natural frequencies is crucial both for sensor network planning, to ensure optimal placement, and for operation, to detect frequency shifts that may indicate structural damage. However, traditional frequency detection methods, such as peak picking of the Spectrum or Power Spectral Density (PSD), are highly dependent on structural and environmental conditions. In highly variable vibrational environments, such as cultural heritage sites, stadiums, and transportation hubs, these methods often prove inadequate, leading to false modal identification. This study applies coherence analysis to vibrational measurements as a more reliable alternative that overcomes the limitations of traditional frequency extraction techniques. To evaluate its effectiveness, Magnitude-Squared Coherence (MSC), Squared Cross-Spectrum (SCS), and Wavelet Coherence (WC) were tested and compared with PSD analysis. Vibrational data were collected from a sensor network deployed at the Civil Museum of Castello Ursino (Catania, Italy), a site characterized by high structural complexity and variable visitor-induced vibrations. Results demonstrate that coherence analysis surpasses the limitations of traditional frequency identification techniques, with SCS and WC outperforming MSC in distinguishing resonance frequencies and providing a more stable and reliable frequency estimation. This approach enhances sensor network design by improving frequency detection, ensuring data reliability, and supporting long-term monitoring through instrumental drift detection, thus strengthening structural health monitoring in heritage sites. Full article

(This article belongs to the Section Actuators, Sensors and Devices)

► Show Figures

Figure 1

17 pages, 1797 KiB

Open AccessArticle

Advanced Diagnosis of Cardiac and Respiratory Diseases from Chest X-Ray Imagery Using Deep Learning Ensembles

by Hemal Nakrani, Essa Q. Shahra, Shadi Basurra, Rasheed Mohammad, Edlira Vakaj and Waheb A. Jabbar

J. Sens. Actuator Netw. 2025, 14(2), 44; https://doi.org/10.3390/jsan14020044 - 18 Apr 2025

Abstract

Chest X-ray interpretation is essential for diagnosing cardiac and respiratory diseases. This study introduces a deep learning ensemble approach that integrates Convolutional Neural Networks (CNNs), including ResNet-152, VGG19, EfficientNet, and a Vision Transformer (ViT), to enhance diagnostic accuracy. Using the NIH Chest X-ray [...] Read more.

Chest X-ray interpretation is essential for diagnosing cardiac and respiratory diseases. This study introduces a deep learning ensemble approach that integrates Convolutional Neural Networks (CNNs), including ResNet-152, VGG19, EfficientNet, and a Vision Transformer (ViT), to enhance diagnostic accuracy. Using the NIH Chest X-ray dataset, the methodology involved comprehensive preprocessing, data augmentation, and model optimization techniques to address challenges such as label imbalance and feature variability. Among the individual models, VGG19 exhibited strong performance with a Hamming Loss of 0.1335 and high accuracy in detecting Edema, while ViT excelled in classifying certain conditions like Hernia. Despite the strengths of individual models, the ensemble meta-model achieved the best overall performance, with a Hamming Loss of 0.1408 and consistently higher ROC-AUC values across multiple diseases, demonstrating its superior capability to handle complex classification tasks. This robust ensemble learning framework underscores its potential for reliable and precise disease detection, offering significant improvements over traditional methods. The findings highlight the value of integrating diverse model architectures to address the complexities of multi-label chest X-ray classification, providing a pathway for more accurate, scalable, and accessible diagnostic tools in clinical practice. Full article

► Show Figures

Figure 1

25 pages, 2027 KiB

Open AccessArticle

Priority-Based Data Flow Control for Long-Range Wide Area Networks in Internet of Military Things

by Rachel Kufakunesu, Herman C. Myburgh and Allan De Freitas

J. Sens. Actuator Netw. 2025, 14(2), 43; https://doi.org/10.3390/jsan14020043 - 16 Apr 2025

Abstract

The Internet of Military Things (IoMT) is transforming defense operations by enabling the seamless integration of sensors and actuators for the real-time transmission of critical data in diverse military environments. End devices (EDs) collect essential information, including troop locations, health metrics, equipment status, [...] Read more.

The Internet of Military Things (IoMT) is transforming defense operations by enabling the seamless integration of sensors and actuators for the real-time transmission of critical data in diverse military environments. End devices (EDs) collect essential information, including troop locations, health metrics, equipment status, and environmental conditions, which are processed to enhance situational awareness and operational efficiency. In scenarios involving large-scale deployments across remote or austere regions, wired communication systems are often impractical and cost-prohibitive. Wireless sensor networks (WSNs) provide a cost-effective alternative, with Long-Range Wide Area Network (LoRaWAN) emerging as a leading protocol due to its extensive coverage, low energy consumption, and reliability. Existing LoRaWAN network simulation modules, such as those in ns-3, primarily support uniform periodic data transmissions, limiting their applicability in critical military and healthcare contexts that demand adaptive transmission rates, resource optimization, and prioritized data delivery. These limitations are particularly pronounced in healthcare monitoring, where frequent, high-rate data transmission is vital but can strain the network’s capacity. To address these challenges, we developed an enhanced sensor data sender application capable of simulating priority-based traffic within LoRaWAN, specifically targeting use cases like border security and healthcare monitoring. This study presents a priority-based data flow control protocol designed to optimize network performance under high-rate healthcare data conditions while maintaining overall system reliability. Simulation results demonstrate that the proposed protocol effectively mitigates performance bottlenecks, ensuring robust and energy-efficient communication in critical IoMT applications within austere environments. Full article

(This article belongs to the Special Issue Security and Smart Applications in IoT and Wireless Sensor and Actuator Networks)

► Show Figures

Figure 1

22 pages, 1052 KiB

Open AccessArticle

Enhancing Sensor-Based Human Physical Activity Recognition Using Deep Neural Networks

by Minyar Sassi Hidri, Adel Hidri, Suleiman Ali Alsaif, Muteeb Alahmari and Eman AlShehri

J. Sens. Actuator Netw. 2025, 14(2), 42; https://doi.org/10.3390/jsan14020042 - 14 Apr 2025

Abstract

Human activity recognition (HAR) is the task of classifying sequences of data into defined movements. Taking advantage of deep learning (DL) methods, this research investigates and optimizes neural network architectures to effectively classify physical activities from smartphone accelerometer data. Unlike traditional machine learning [...] Read more.

Human activity recognition (HAR) is the task of classifying sequences of data into defined movements. Taking advantage of deep learning (DL) methods, this research investigates and optimizes neural network architectures to effectively classify physical activities from smartphone accelerometer data. Unlike traditional machine learning (ML) methods employing manually crafted features, our approach employs automated feature learning with three deep learning architectures: Convolutional Neural Networks (CNN), CNN-based autoencoders, and Long Short-Term Memory Recurrent Neural Networks (LSTM RNN). The contribution of this work is primarily in optimizing LSTM RNN to leverage the most out of temporal relationships between sensor data, significantly improving classification accuracy. Experimental outcomes for the WISDM dataset show that the proposed LSTM RNN model achieves 96.1% accuracy, outperforming CNN-based approaches and current ML-based methods. Compared to current works, our optimized frameworks achieve up to 6.4% higher classification performance, which means that they are more appropriate for real-time HAR. Full article

► Show Figures

Figure 1

14 pages, 2548 KiB

Open AccessArticle

Lossless Compression with Trie-Based Shared Dictionary for Omics Data in Edge–Cloud Frameworks

by Rani Adam, Daniel R. Catchpoole, Simeon J. Simoff, Zhonglin Qu, Paul J. Kennedy and Quang Vinh Nguyen

J. Sens. Actuator Netw. 2025, 14(2), 41; https://doi.org/10.3390/jsan14020041 - 9 Apr 2025

Abstract

The growing complexity and volume of genomic and omics data present critical challenges for storage, transfer, and analysis in edge–cloud platforms. Existing compression techniques often involve trade-offs between efficiency and speed, requiring innovative approaches that ensure scalability and cost-effectiveness. This paper introduces a [...] Read more.

The growing complexity and volume of genomic and omics data present critical challenges for storage, transfer, and analysis in edge–cloud platforms. Existing compression techniques often involve trade-offs between efficiency and speed, requiring innovative approaches that ensure scalability and cost-effectiveness. This paper introduces a lossless compression method that integrates Trie-based shared dictionaries within an edge–cloud architecture. It presents a software-centric scientific research process of the design and evaluation of the proposed compression method. By enabling localized preprocessing at the edge, our approach reduces data redundancy before cloud transmission, thereby optimizing both storage and network efficiency. A global shared dictionary is constructed using N-gram analysis to identify and prioritize repeated sequences across multiple files. A lightweight index derived from this dictionary is then pushed to edge nodes, where Trie-based sequence replacement is applied to eliminate redundancy locally. The preprocessed data are subsequently transmitted to the cloud, where advanced compression algorithms, such as Zstd, GZIP, Snappy, and LZ4, further compress them. Evaluation on real patient omics datasets from B-cell Acute Lymphoblastic Leukemia (B-ALL) and Chronic Lymphocytic Leukemia (CLL) demonstrates that edge preprocessing significantly improves compression ratios, reduces upload times, and enhances scalability in hybrid cloud frameworks. Full article

► Show Figures

Figure 1

17 pages, 3162 KiB

Open AccessArticle

Deepfake Image Classification Using Decision (Binary) Tree Deep Learning

by Mariam Alrajeh and Aida Al-Samawi

J. Sens. Actuator Netw. 2025, 14(2), 40; https://doi.org/10.3390/jsan14020040 - 8 Apr 2025

Abstract

The unprecedented rise of deepfake technologies, leveraging sophisticated AI like Generative Adversarial Networks (GANs) and diffusion-based models, presents both opportunities and challenges in terms of digital media authenticity. In response, this study introduces a novel deep neural network ensemble that utilizes a tree-based [...] Read more.

The unprecedented rise of deepfake technologies, leveraging sophisticated AI like Generative Adversarial Networks (GANs) and diffusion-based models, presents both opportunities and challenges in terms of digital media authenticity. In response, this study introduces a novel deep neural network ensemble that utilizes a tree-based hierarchical architecture integrating a vision transformer, ResNet, EfficientNet, and DenseNet to address the pressing need for effective deepfake detection. Our model exhibits a high degree of adaptability across varied datasets and demonstrates state-of-the-art performance, achieving up to 97.25% accuracy and a weighted F1 score of 97.28%. By combining the strengths of various convolutional networks and the vision transformer, our approach underscores a scalable solution for mitigating the risks associated with manipulated media. Full article

(This article belongs to the Section Network Security and Privacy)

► Show Figures

Figure 1

24 pages, 22704 KiB

Open AccessReview

Urban Air Mobility, Personal Drones, and the Safety of Occupants—A Comprehensive Review

by Dmytro Zhyriakov, Mariusz Ptak and Marek Sawicki

J. Sens. Actuator Netw. 2025, 14(2), 39; https://doi.org/10.3390/jsan14020039 - 6 Apr 2025

Abstract

Urban air mobility (UAM) is expected to provide environmental benefits while enhancing transportation for citizens and businesses, particularly in commercial and emergency medical applications. The rapid development of electric vertical take-off and landing (eVTOL) aircraft has demonstrated the potential to introduce new technological [...] Read more.

Urban air mobility (UAM) is expected to provide environmental benefits while enhancing transportation for citizens and businesses, particularly in commercial and emergency medical applications. The rapid development of electric vertical take-off and landing (eVTOL) aircraft has demonstrated the potential to introduce new technological capabilities to the market, fostering visions of widespread and diverse UAM applications. This paper reviews state-of-the-art occupant safety for personal drones and examines existing occupant protection methods in the aircraft. The study serves as a guide for stakeholders, including regulators, manufacturers, researchers, policymakers, and industry professionals—by providing insights into the regulatory landscape and safety assurance frameworks for eVTOL aircraft in UAM applications. Furthermore, we present a functional hazard assessment (FHA) conducted on a reference concept, detailing the process, decision-making considerations, and key variations. The analysis illustrates the FHA methodology while discussing the trade-offs involved in safety evaluations. Additionally, we provide a summary and a featured description of current eVTOL aircraft, highlighting their key characteristics and technological advancements. Full article

(This article belongs to the Special Issue Advances in Intelligent Transportation Systems (ITS))

► Show Figures

Figure 1

33 pages, 9824 KiB

Open AccessArticle

An Efficient Framework for Peer Selection in Dynamic P2P Network Using Q Learning with Fuzzy Linear Programming

by Mahalingam Anandaraj, Tahani Albalawi and Mohammad Alkhatib

J. Sens. Actuator Netw. 2025, 14(2), 38; https://doi.org/10.3390/jsan14020038 - 2 Apr 2025

Abstract

This paper proposes a new approach to integrating Q learning into the fuzzy linear programming (FLP) paradigm to improve peer selection in P2P networks. Using Q learning, the proposed method employs real-time feedback to adjust and update peer selection policies. The FLP framework [...] Read more.

This paper proposes a new approach to integrating Q learning into the fuzzy linear programming (FLP) paradigm to improve peer selection in P2P networks. Using Q learning, the proposed method employs real-time feedback to adjust and update peer selection policies. The FLP framework enriches this process by dealing with imprecise information through fuzzy logic. It is used to achieve multiple objectives, such as enhancing the throughput rate, reducing the delay, and guaranteeing a reliable connection. This integration effectively solves the problem of network uncertainty, making the network configuration more stable and flexible. It is also important to note that throughout the use of the Q-learning agent in the network, various state metric indicators, including available bandwidth, latency, packet drop rates, and connectivity of nodes, are observed and recorded. It then selects actions by choosing optimal peers for each node and updating a Q table that defines states and actions based on these performance indices. This reward system guides the agent’s learning, refining its peer selection policy over time. The FLP framework supports the Q-learning agent by providing optimized solutions that balance conflicting objectives under uncertain conditions. Fuzzy parameters capture variability in network metrics, and the FLP model solves a fuzzy linear programming problem, offering guidelines for the Q-learning agent’s decisions. The proposed method is evaluated under different experimental settings to reveal its effectiveness. The Erdos–Renyi model simulation is used, and it shows that throughput increased by 21% and latency decreased by 40%. The computational efficiency was also notably improved, with computation times diminishing by up to five orders of magnitude compared to traditional methods. Full article

► Show Figures

Figure 1

42 pages, 1685 KiB

Open AccessArticle

Heterogeneity Challenges of Federated Learning for Future Wireless Communication Networks

by Lorena Isabel Barona López and Thomás Borja Saltos

J. Sens. Actuator Netw. 2025, 14(2), 37; https://doi.org/10.3390/jsan14020037 - 1 Apr 2025

Abstract

Two technologies of great interest in recent years—Artificial Intelligence (AI) and massive wireless communication networks—have found a significant point of convergence through Federated Learning (FL). Federated Learning is a Machine Learning (ML) technique that enables multiple participants to collaboratively train a model while [...] Read more.

Two technologies of great interest in recent years—Artificial Intelligence (AI) and massive wireless communication networks—have found a significant point of convergence through Federated Learning (FL). Federated Learning is a Machine Learning (ML) technique that enables multiple participants to collaboratively train a model while keeping their data local. Several studies indicate that while improving performance metrics—such as accuracy, loss reduction, or computation time—is a primary goal, achieving this in real-world scenarios remains challenging. This difficulty arises due to various heterogeneity characteristics inherent to the wireless devices participating in the Federation. Heterogeneity in Federated Learning arises when participants contribute differently, leading to challenges in the model training process. Heterogeneity in Federated Learning may appear in architecture, statistics, and behavior. System heterogeneity arises from differences in device capabilities, including processing power, transmission speeds, availability, energy constraints, and network limitations, among others. Statistical heterogeneity occurs when participants contribute non-independent and non-identically distributed (non-IID) data. This situation can harm the global model instead of improving it, especially when the data are of poor quality or too scarce. The third type, behavioral heterogeneity, refers to cases where participants are unwilling to engage or expect rewards despite minimal effort. Given the growing research in this area, we present a summary of heterogeneity characteristics in Federated Learning to provide a broader perspective on this emerging technology. We also outline key challenges, opportunities, and future directions for Federated Learning. Finally, we conduct a simulation using the LEAF framework to illustrate the impact of heterogeneity in Federated Learning. Full article

(This article belongs to the Special Issue Federated Learning: Applications and Future Directions)

► Show Figures

Figure 1

36 pages, 28868 KiB

Open AccessArticle

Lower-Complexity Multi-Layered Security Partitioning Algorithm Based on Chaos Mapping-DWT Transform for WA/SNs

by Tarek Srour, Mohsen A. M. El-Bendary, Mostafa Eltokhy, Atef E. Abouelazm, Ahmed A. F. Youssef and Ali M. El-Rifaie

J. Sens. Actuator Netw. 2025, 14(2), 36; https://doi.org/10.3390/jsan14020036 - 31 Mar 2025

Abstract

The resource limitations of Low-Power Wireless Networks (LP-WNs), such as Wireless Sensor Networks (WSNs), Wireless Actuator/Sensor Networks (WA/SNs), and Internet of Things (IoT) outdoor applications, restrict the utilization of the error-performance-enhancing techniques and the use of the powerful and robust security tools. Therefore, [...] Read more.

The resource limitations of Low-Power Wireless Networks (LP-WNs), such as Wireless Sensor Networks (WSNs), Wireless Actuator/Sensor Networks (WA/SNs), and Internet of Things (IoT) outdoor applications, restrict the utilization of the error-performance-enhancing techniques and the use of the powerful and robust security tools. Therefore, these LP-WN applications require special techniques to satisfy the requirements of a low data loss rate and satisfy the security requirements while considering the accepted level of complexity and power efficiency of these techniques. This paper focuses on proposing a power-efficient, robust cryptographic algorithm for the WA/SNs. The lower-complexity cryptographic algorithm is proposed, based on merging the data composition tools utilizing data transforms and chaos mapping techniques. The decomposing tool is performed by the various data transforms: Discrete Cosine Transform (DCT), Discrete Cosine Wavelet (DWT), Fast Fourier Transform (FFT), and Walsh Hadamard Transform (WHT); the DWT performs better with efficient complexity. It is utilized to separate the plaintext into the main portion and side information portions to reduce more than 50% of complexity. The main plaintext portion is ciphered in the series of cryptography to reduce the complexity and increase the security capabilities of the proposed algorithm by two chaos mappings. The process of reduction saves complexity and is employed to feed the series of chaos cryptography without increasing the complexity. The two chaos mappings are used, and two-dimensional (2D) chaos logistic maps are used due to their high sensitivity to noise and attacks. The chaos 2D baker map is utilized due to its high secret key managing flexibility and high sensitivity to initial conditions and plaintext dimensions. Several computer experiments are demonstrated to evaluate the robustness, reliability, and applicability of the proposed complexity-efficient crypto-system algorithm in the presence of various attacks. The results prove the high suitability of the proposed lower-complexity crypto-system for WASN and LP-WN applications due to its robustness in the presence of attacks and its power efficiency. Full article

(This article belongs to the Special Issue Applications of Wireless Sensor Networks: Innovations and Future Trends)

► Show Figures

Figure 1

Journal Description

Journal of Sensor and Actuator Networks

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Further Information

Guidelines

MDPI Initiatives

Follow MDPI