Previous Issue
Volume 14, June
 
 

J. Sens. Actuator Netw., Volume 14, Issue 4 (August 2025) – 22 articles

Cover Story (view full-size image): Images often contain sensitive content such as personal data, medical records, confidential documents, or military intelligence. Unauthorized access can lead to severe consequences. Encryption is an effective approach to safeguarding image security and privacy. Recently, a JPEG image encryption method using an adaptive encryption key was presented by He et al. However, He’s scheme has shown to be vulnerable to chosen plaintext attack. This study first demonstrates that the adaptive key generation in He’s scheme introduces exploitable security risks and then proposes an improved chosen plaintext attack scheme by replacing the single-permutation approach with an additive value method, which can significantly improve the attack success rates. The proposed method also provides insights for developing more robust image cryptographic schemes. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Select all
Export citation of selected articles as:
2 pages, 139 KiB  
Correction
Correction: Japertas et al. Experimental Study of Lidar System for a Static Object in Adverse Weather Conditions. J. Sens. Actuator Netw. 2025, 14, 56
by Saulius Japertas, Rūta Jankūnienė and Roy Knechtel
J. Sens. Actuator Netw. 2025, 14(4), 85; https://doi.org/10.3390/jsan14040085 - 15 Aug 2025
Viewed by 90
Abstract
There was an error in the original publication [...] Full article
23 pages, 3579 KiB  
Article
Loss Clustering at MSP Buffer
by Andrzej Chydzinski and Blazej Adamczyk
J. Sens. Actuator Netw. 2025, 14(4), 84; https://doi.org/10.3390/jsan14040084 - 13 Aug 2025
Viewed by 214
Abstract
Packet losses cause a decline in the performance of packet networks, and this decline is related not only to the percentage of losses but also to the clustering of them together in series. We study how the correlation of packet sizes influences this [...] Read more.
Packet losses cause a decline in the performance of packet networks, and this decline is related not only to the percentage of losses but also to the clustering of them together in series. We study how the correlation of packet sizes influences this clustering when the losses are caused by buffer overflows. Specifically, for a model of a buffer with correlated packet sizes, we derive the burst ratio parameter, an intuitive metric for the inclination of losses to cluster. In addition to the burst ratio, we obtain the sequential losses distribution in the first, k-th, and stationary overflow periods. The Markovian Service Process (MSP) used by the model empowers it to mimic arbitrary packet size distributions and arbitrary correlation strengths. Using numeric examples, the impact of packet size correlation, buffer size, and traffic intensity on the burst ratio is showcased and discussed. Full article
(This article belongs to the Section Communications and Networking)
Show Figures

Figure 1

23 pages, 1337 KiB  
Article
Can Differential Privacy Hinder Poisoning Attack Detection in Federated Learning?
by Chaitanya Aggarwal, Divya G. Nair, Jafar Aco Mohammadi, Jyothisha J. Nair and Jörg Ott
J. Sens. Actuator Netw. 2025, 14(4), 83; https://doi.org/10.3390/jsan14040083 - 6 Aug 2025
Viewed by 449
Abstract
We consider the problem of data poisoning attack detection in a federated learning (FL) setup with differential privacy (DP). Local DP in FL ensures that privacy leakage caused by shared gradients is controlled by adding randomness to the process. We are interested in [...] Read more.
We consider the problem of data poisoning attack detection in a federated learning (FL) setup with differential privacy (DP). Local DP in FL ensures that privacy leakage caused by shared gradients is controlled by adding randomness to the process. We are interested in studying the effect of the Gaussian mechanism in the detection of different data poisoning attacks. As the additive noise from DP could hide poisonous data, the effectiveness of detection algorithms should be analyzed. We present two poisonous data detection algorithms and one malicious client identification algorithm. For the latter, we show that the effect of DP noise decreases as the size of the neural network increases. We further demonstrate this effect alongside the performance of these algorithms on three publicly available datasets. Full article
(This article belongs to the Special Issue Federated Learning: Applications and Future Directions)
Show Figures

Figure 1

18 pages, 1239 KiB  
Article
A Digitally Controlled Adaptive Current Interface for Accurate Measurement of Resistive Sensors in Embedded Sensing Systems
by Jirapong Jittakort and Apinan Aurasopon
J. Sens. Actuator Netw. 2025, 14(4), 82; https://doi.org/10.3390/jsan14040082 - 4 Aug 2025
Viewed by 381
Abstract
This paper presents a microcontroller-based technique for accurately measuring resistive sensors over a wide dynamic range using an adaptive constant current source. Unlike conventional voltage dividers or fixed-current methods—often limited by reduced resolution and saturation when sensor resistance varies across several decades—the proposed [...] Read more.
This paper presents a microcontroller-based technique for accurately measuring resistive sensors over a wide dynamic range using an adaptive constant current source. Unlike conventional voltage dividers or fixed-current methods—often limited by reduced resolution and saturation when sensor resistance varies across several decades—the proposed system dynamically adjusts the excitation current to maintain optimal Analog-to-Digital Converter (ADC) input conditions. The measurement circuit employs a fixed reference resistor and an inverting amplifier configuration, where the excitation current is generated by one or more pulse-width modulated (PWM) signals filtered through low-pass RC networks. A microcontroller selects the appropriate PWM channel to ensure that the output voltage remains within the ADC’s linear range. To support multiple sensors, an analog switch enables sequential measurements using the same dual-PWM current source. The full experimental implementation uses four op-amps to support modularity, buffering, and dual-range operation. Experimental results show accurate measurement of resistances from 1 kΩ to 100 kΩ, with maximum relative errors of 0.15% in the 1–10 kΩ range and 0.33% in the 10–100 kΩ range. The method provides a low-cost, scalable, and digitally controlled solution suitable for embedded resistive sensing applications without the need for high-resolution ADCs or programmable gain amplifiers. Full article
(This article belongs to the Section Actuators, Sensors and Devices)
Show Figures

Figure 1

20 pages, 2223 KiB  
Article
Category Attribute-Oriented Heterogeneous Resource Allocation and Task Offloading for SAGIN Edge Computing
by Yuan Qiu, Xiang Luo, Jianwei Niu, Xinzhong Zhu and Yiming Yao
J. Sens. Actuator Netw. 2025, 14(4), 81; https://doi.org/10.3390/jsan14040081 - 1 Aug 2025
Viewed by 364
Abstract
Space-Air-Ground Integrated Network (SAGIN), which is considered a network architecture with great development potential, exhibits significant cross-domain collaboration characteristics at present. However, most of the existing works ignore the matching and adaptability of differential tasks and heterogeneous resources, resulting in significantly inefficient task [...] Read more.
Space-Air-Ground Integrated Network (SAGIN), which is considered a network architecture with great development potential, exhibits significant cross-domain collaboration characteristics at present. However, most of the existing works ignore the matching and adaptability of differential tasks and heterogeneous resources, resulting in significantly inefficient task execution and undesirable network performance. As a consequence, we formulate a category attribute-oriented resource allocation and task offloading optimization problem with the aim of minimizing the overall scheduling cost. We first introduce a task–resource matching matrix to facilitate optimal task offloading policies with computation resources. In addition, virtual queues are constructed to take the impacts of randomized task arrival into account. To solve the optimization objective which jointly considers bandwidth allocation, transmission power control and task offloading decision effectively, we proposed a deep reinforcement learning (DRL) algorithm framework considering type matching. Simulation experiments demonstrate the effectiveness of our proposed algorithm as well as superior performance compared to others. Full article
(This article belongs to the Section Communications and Networking)
Show Figures

Figure 1

21 pages, 4688 KiB  
Article
Nondestructive Inspection of Steel Cables Based on YOLOv9 with Magnetic Flux Leakage Images
by Min Zhao, Ning Ding, Zehao Fang, Bingchun Jiang, Jiaming Zhong and Fuqin Deng
J. Sens. Actuator Netw. 2025, 14(4), 80; https://doi.org/10.3390/jsan14040080 - 1 Aug 2025
Viewed by 429
Abstract
The magnetic flux leakage (MFL) method is widely acknowledged as a highly effective non-destructive evaluation (NDE) technique for detecting local damage in ferromagnetic structures such as steel wire ropes. In this study, a multi-channel MFL sensor module was developed, incorporating a purpose-designed Hall [...] Read more.
The magnetic flux leakage (MFL) method is widely acknowledged as a highly effective non-destructive evaluation (NDE) technique for detecting local damage in ferromagnetic structures such as steel wire ropes. In this study, a multi-channel MFL sensor module was developed, incorporating a purpose-designed Hall sensor array and magnetic yokes specifically shaped for steel cables. To validate the proposed damage detection method, artificial damages of varying degrees were inflicted on wire rope specimens through experimental testing. The MFL sensor module facilitated the scanning of the damaged specimens and measurement of the corresponding MFL signals. In order to improve the signal-to-noise ratio, a comprehensive set of signal processing steps, including channel equalization and normalization, was implemented. Subsequently, the detected MFL distribution surrounding wire rope defects was transformed into MFL images. These images were then analyzed and processed utilizing an object detection method, specifically employing the YOLOv9 network, which enables accurate identification and localization of defects. Furthermore, a quantitative defect detection method based on image size was introduced, which is effective for quantifying defects using the dimensions of the anchor frame. The experimental results demonstrated the effectiveness of the proposed approach in detecting and quantifying defects in steel cables, which combines deep learning-based analysis of MFL images with the non-destructive inspection of steel cables. Full article
Show Figures

Figure 1

40 pages, 1548 KiB  
Article
Real-Time Service Migration in Edge Networks: A Survey
by Yutong Zhang, Ke Zhao, Yihong Yang and Zhangbing Zhou
J. Sens. Actuator Netw. 2025, 14(4), 79; https://doi.org/10.3390/jsan14040079 - 1 Aug 2025
Viewed by 691
Abstract
With the rapid proliferation of Internet of Things (IoT) devices and mobile applications and the growing demand for low-latency services, edge computing has emerged as a transformative paradigm that brings computation and storage closer to end users. However, [...] Read more.
With the rapid proliferation of Internet of Things (IoT) devices and mobile applications and the growing demand for low-latency services, edge computing has emerged as a transformative paradigm that brings computation and storage closer to end users. However, the dynamic nature and limited resources of edge networks bring challenges such as load imbalance and high latency while satisfying user requests. Service migration, the dynamic redeployment of service instances across distributed edge nodes, has become a key enabler for solving these challenges and optimizing edge network characteristics. Moreover, the low-latency nature of edge computing requires that service migration strategies must be in real time in order to ensure latency requirements. Thus, this paper presents a systematic survey of real-time service migration in edge networks. Specifically, we first introduce four network architectures and four basic models for real-time service migration. We then summarize four research motivations for real-time service migration and the real-time guarantee introduced during the implementation of migration strategies. To support these motivations, we present key techniques for solving the task of real-time service migration and how these algorithms and models facilitate the real-time performance of migration. We also explore latency-sensitive application scenarios, such as smart cities, smart homes, and smart manufacturing, where real-time service migration plays a critical role in sustaining performance and adaptability under dynamic conditions. Finally, we summarize the key challenges and outline promising future research directions for real-time service migration. This survey aims to provide a structured and in-depth theoretical foundation to guide future research on real-time service migration in edge networks. Full article
Show Figures

Figure 1

24 pages, 2151 KiB  
Article
Federated Learning-Based Intrusion Detection in IoT Networks: Performance Evaluation and Data Scaling Study
by Nurtay Albanbay, Yerlan Tursynbek, Kalman Graffi, Raissa Uskenbayeva, Zhuldyz Kalpeyeva, Zhastalap Abilkaiyr and Yerlan Ayapov
J. Sens. Actuator Netw. 2025, 14(4), 78; https://doi.org/10.3390/jsan14040078 - 23 Jul 2025
Viewed by 1334
Abstract
This paper presents a large-scale empirical study aimed at identifying the optimal local deep learning model and data volume for deploying intrusion detection systems (IDS) on resource-constrained IoT devices using federated learning (FL). While previous studies on FL-based IDS for IoT have primarily [...] Read more.
This paper presents a large-scale empirical study aimed at identifying the optimal local deep learning model and data volume for deploying intrusion detection systems (IDS) on resource-constrained IoT devices using federated learning (FL). While previous studies on FL-based IDS for IoT have primarily focused on maximizing accuracy, they often overlook the computational limitations of IoT hardware and the feasibility of local model deployment. In this work, three deep learning architectures—a deep neural network (DNN), a convolutional neural network (CNN), and a hybrid CNN+BiLSTM—are trained using the CICIoT2023 dataset within a federated learning environment simulating up to 150 IoT devices. The study evaluates how detection accuracy, convergence speed, and inference costs (latency and model size) vary across different local data scales and model complexities. Results demonstrate that CNN achieves the best trade-off between detection performance and computational efficiency, reaching ~98% accuracy with low latency and a compact model footprint. The more complex CNN+BiLSTM architecture yields slightly higher accuracy (~99%) at a significantly greater computational cost. Deployment tests on Raspberry Pi 5 devices confirm that all three models can be effectively implemented on real-world IoT edge hardware. These findings offer practical guidance for researchers and practitioners in selecting scalable and lightweight IDS models suitable for real-world federated IoT deployments, supporting secure and efficient anomaly detection in urban IoT networks. Full article
(This article belongs to the Special Issue Federated Learning: Applications and Future Directions)
Show Figures

Figure 1

16 pages, 1655 KiB  
Article
FO-DEMST: Optimized Multi-Scale Transformer with Dual-Encoder Architecture for Feeding Amount Prediction in Sea Bass Aquaculture
by Hongpo Wang, Qihui Zhang, Hong Zhou, Yunchen Tian, Yongcheng Jiang and Jianing Quan
J. Sens. Actuator Netw. 2025, 14(4), 77; https://doi.org/10.3390/jsan14040077 - 22 Jul 2025
Viewed by 387
Abstract
Traditional methods for predicting feeding amounts rely on historical data and experience but fail to account for non-linear fish growth and the influence of water quality and meteorological factors. This study presents a novel approach for sea bass feeding prediction based on Spearman [...] Read more.
Traditional methods for predicting feeding amounts rely on historical data and experience but fail to account for non-linear fish growth and the influence of water quality and meteorological factors. This study presents a novel approach for sea bass feeding prediction based on Spearman + RF feature optimization and multi-scale feature fusion using a transformer model. A logistic growth curve model is used to analyze sea bass growth and establish the relationship between biomass and feeding amount. Spearman correlation analysis and random forest optimize the feature set for improved prediction accuracy. A dual-encoder structure incorporates historical feeding data and biomass along with water quality and meteorological information. Multi-scale feature fusion addresses time-scale inconsistencies between input variables The results showed that the MSE and MAE of the improved transformer model for sea bass feeding prediction were 0.42 and 0.31, respectively, which decreased by 43% in MSE and 33% in MAE compared to the traditional transformer model. Full article
(This article belongs to the Special Issue Remote Sensing and IoT Application for Smart Agriculture)
Show Figures

Figure 1

40 pages, 17591 KiB  
Article
Research and Education in Robotics: A Comprehensive Review, Trends, Challenges, and Future Directions
by Mutaz Ryalat, Natheer Almtireen, Ghaith Al-refai, Hisham Elmoaqet and Nathir Rawashdeh
J. Sens. Actuator Netw. 2025, 14(4), 76; https://doi.org/10.3390/jsan14040076 - 16 Jul 2025
Viewed by 1876
Abstract
Robotics has emerged as a transformative discipline at the intersection of the engineering, computer science, and cognitive sciences. This state-of-the-art review explores the current trends, methodologies, and challenges in both robotics research and education. This paper presents a comprehensive review of the evolution [...] Read more.
Robotics has emerged as a transformative discipline at the intersection of the engineering, computer science, and cognitive sciences. This state-of-the-art review explores the current trends, methodologies, and challenges in both robotics research and education. This paper presents a comprehensive review of the evolution of robotics, tracing its development from early automation to intelligent, autonomous systems. Key enabling technologies, such as Artificial Intelligence (AI), soft robotics, the Internet of Things (IoT), and swarm intelligence, are examined along with real-world applications in healthcare, manufacturing, agriculture, and sustainable smart cities. A central focus is placed on robotics education, where hands-on, interdisciplinary learning is reshaping curricula from K–12 to postgraduate levels. This paper analyzes instructional models including project-based learning, laboratory work, capstone design courses, and robotics competitions, highlighting their effectiveness in developing both technical and creative competencies. Widely adopted platforms such as the Robot Operating System (ROS) are briefly discussed in the context of their educational value and real-world alignment. Through case studies, institutional insights, and synthesis of academic and industry practices, this review underscores the vital role of robotics education in fostering innovation, systems thinking, and workforce readiness. The paper concludes by identifying the key challenges and future directions to guide researchers, educators, industry stakeholders, and policymakers in advancing robotics as both technological and educational frontiers. Full article
Show Figures

Figure 1

20 pages, 7588 KiB  
Article
Dual-Purpose Star Tracker and Space Debris Detector: Miniature Instrument for Small Satellites
by Beltran N. Arribas, João G. Maia, João P. Castanheira, Joel Filho, Rui Melicio, Hugo Onderwater, Paulo Gordo, R. Policarpo Duarte and André R. R. Silva
J. Sens. Actuator Netw. 2025, 14(4), 75; https://doi.org/10.3390/jsan14040075 - 16 Jul 2025
Viewed by 823
Abstract
This paper presents the conception, design and real miniature instrument implementation of a dual-purpose sensor for small satellites that can act as a star tracker and space debris detector. In the previous research work, the authors conceived, designed and implemented a breadboard consisting [...] Read more.
This paper presents the conception, design and real miniature instrument implementation of a dual-purpose sensor for small satellites that can act as a star tracker and space debris detector. In the previous research work, the authors conceived, designed and implemented a breadboard consisting of a computer laptop, a camera interface and camera controller, an image sensor, an optics system, a temperature sensor and a temperature controller. It showed that the instrument was feasible. In this paper, a new real star tracker miniature instrument is designed, physically realized and tested. The implementation follows a New Space approach; it is made with Commercial Off-the-Shelf (COTS) components with space heritage. The instrument’s development, implementation and testing are presented. Full article
Show Figures

Figure 1

23 pages, 6850 KiB  
Article
Optimizing Energy Consumption in Public Institutions Using AI-Based Load Shifting and Renewable Integration
by Otilia Elena Dragomir, Florin Dragomir and Marius Păun
J. Sens. Actuator Netw. 2025, 14(4), 74; https://doi.org/10.3390/jsan14040074 - 15 Jul 2025
Viewed by 524
Abstract
This paper details the development and implementation of an intelligent energy efficiency system for an electrical grid that incorporates renewable energy sources, specifically photovoltaic systems. The system is applied in a small locality of approximately 8000 inhabitants and aims to optimize energy consumption [...] Read more.
This paper details the development and implementation of an intelligent energy efficiency system for an electrical grid that incorporates renewable energy sources, specifically photovoltaic systems. The system is applied in a small locality of approximately 8000 inhabitants and aims to optimize energy consumption in public institutions by scheduling electrical appliances during periods of surplus PV energy production. The proposed solution employs a hybrid neuro-fuzzy approach combined with scheduling techniques to intelligently shift loads and maximize the use of locally generated green energy. This enables appliances, particularly schedulable and schedulable non-interruptible ones, to operate during peak PV production hours, thereby minimizing reliance on the national grid and improving overall energy efficiency. This directly reduces the cost of electricity consumption from the national grid. Furthermore, a comprehensive power quality analysis covering variables including harmonic distortion and voltage stability is proposed. The results indicate that while photovoltaic systems, being switching devices, can introduce some harmonic distortion, particularly during peak inverter operation or transient operating regimes, and flicker can exceed standard limits during certain periods, the overall voltage quality is maintained if proper inverter controls and grid parameters are adhered to. The system also demonstrates potential for scalability and integration with energy storage systems for enhanced future performance. Full article
(This article belongs to the Section Network Services and Applications)
Show Figures

Figure 1

15 pages, 2538 KiB  
Article
Parallel Eclipse-Aware Routing on FPGA for SpaceWire-Based OBC in LEO Satellite Networks
by Jin Hyung Park, Heoncheol Lee and Myonghun Han
J. Sens. Actuator Netw. 2025, 14(4), 73; https://doi.org/10.3390/jsan14040073 - 15 Jul 2025
Viewed by 528
Abstract
Low Earth orbit (LEO) satellite networks deliver superior real-time performance and responsiveness compared to conventional satellite networks, despite technical and economic challenges such as high deployment costs and operational complexity. Nevertheless, rapid topology changes and severe energy constraints of LEO satellites make real-time [...] Read more.
Low Earth orbit (LEO) satellite networks deliver superior real-time performance and responsiveness compared to conventional satellite networks, despite technical and economic challenges such as high deployment costs and operational complexity. Nevertheless, rapid topology changes and severe energy constraints of LEO satellites make real-time routing a persistent challenge. In this paper, we employ field-programmable gate arrays (FPGAs) to overcome the resource limitations of on-board computers (OBCs) and to manage energy consumption effectively using the Eclipse-Aware Routing (EAR) algorithm, and we implement the K-Shortest Paths (KSP) algorithm directly on the FPGA. Our method first generates multiple routes from the source to the destination using KSP, then selects the optimal path based on energy consumption rate, eclipse duration, and estimated transmission load as evaluated by EAR. In large-scale LEO networks, the computational burden of KSP grows substantially as connectivity data become more voluminous and complex. To enhance performance, we accelerate complex computations in the programmable logic (PL) via pipelining and design a collaborative architecture between the processing system (PS) and PL, achieving approximately a 3.83× speedup compared to a PS-only implementation. We validate the feasibility of the proposed approach by successfully performing remote routing-table updates on the SpaceWire-based SpaceWire Brick MK4 network system. Full article
(This article belongs to the Section Communications and Networking)
Show Figures

Figure 1

19 pages, 3841 KiB  
Article
An Improved Chosen Plaintext Attack on JPEG Encryption
by Junhui He, Kaitian Gu, Yihan Huang, Yue Li and Xiang Chen
J. Sens. Actuator Netw. 2025, 14(4), 72; https://doi.org/10.3390/jsan14040072 - 14 Jul 2025
Viewed by 519
Abstract
Format-compatible encryption can be used to ensure the security and privacy of JPEG images. Recently, a JPEG image encryption method proved to be secure against known plaintext attacks by employing an adaptive encryption key, which depends on the histogram of the number of [...] Read more.
Format-compatible encryption can be used to ensure the security and privacy of JPEG images. Recently, a JPEG image encryption method proved to be secure against known plaintext attacks by employing an adaptive encryption key, which depends on the histogram of the number of non-zero alternating current coefficients (ACC) in Discrete Cosine Transform (DCT) blocks. However, this scheme has been demonstrated to be vulnerable to chosen-plaintext attacks (CPA) based on the run consistency of MCUs (RCM) between the original image and the encrypted image. In this paper, an improved CPA scheme is proposed. The method of incrementing run-length values instead of permutation is utilized to satisfy the uniqueness of run sequences of different minimum coded units (MCUs). The experimental results show that the proposed method can successfully recover the outlines of plaintext images from the encrypted images, even with lower-quality factors. Full article
Show Figures

Figure 1

17 pages, 1820 KiB  
Article
A Federated Learning Architecture for Bird Species Classification in Wetlands
by David Mulero-Pérez, Javier Rodriguez-Juan, Tamai Ramirez-Gordillo, Manuel Benavent-Lledo, Pablo Ruiz-Ponce, David Ortiz-Perez, Hugo Hernandez-Lopez, Anatoli Iarovikov, Jose Garcia-Rodriguez, Esther Sebastián-González, Olamide Jogunola, Segun I. Popoola and Bamidele Adebisi
J. Sens. Actuator Netw. 2025, 14(4), 71; https://doi.org/10.3390/jsan14040071 - 9 Jul 2025
Viewed by 694
Abstract
Federated learning allows models to be trained on edge devices with local data, eliminating the need to share data with a central server. This significantly reduces the amount of data transferred from edge devices to central servers, which is particularly important in rural [...] Read more.
Federated learning allows models to be trained on edge devices with local data, eliminating the need to share data with a central server. This significantly reduces the amount of data transferred from edge devices to central servers, which is particularly important in rural areas with limited bandwidth resources. Despite the potential of federated learning to fine-tune deep learning models using data collected from edge devices in low-resource environments, its application in the field of bird monitoring remains underexplored. This study proposes a federated learning pipeline tailored for bird species classification in wetlands. The proposed approach is based on lightweight convolutional neural networks optimized for use on resource-constrained devices. Since the performance of federated learning is strongly influenced by the models used and the experimental setting, this study conducts a comprehensive comparison of well-known lightweight models such as WideResNet, EfficientNetV2, MNASNet, GoogLeNet and ResNet in different training settings. The results demonstrate the importance of the training setting in federated learning architectures and the suitability of the different models for bird species recognition. This work contributes to the wider application of federated learning in ecological monitoring and highlights its potential to overcome challenges such as bandwidth limitations. Full article
(This article belongs to the Special Issue Federated Learning: Applications and Future Directions)
Show Figures

Figure 1

18 pages, 3283 KiB  
Article
AI-Driven Differentiation and Quantification of Metal Ions Using ITIES Electrochemical Sensors
by Muzammil M. N. Ahmed, Parth Ganeriwala, Anthi Savvidou, Nicholas Breen, Siddhartha Bhattacharyya and Pavithra Pathirathna
J. Sens. Actuator Netw. 2025, 14(4), 70; https://doi.org/10.3390/jsan14040070 - 9 Jul 2025
Viewed by 660
Abstract
Electrochemical sensors, particularly those based on ion transfer at the interface between two immiscible electrolyte solutions (ITIES), offer significant advantages such as high selectivity, ease of fabrication, and cost effectiveness for toxic metal ion detection. However, distinguishing between cyclic voltammograms (CVs) of analytes [...] Read more.
Electrochemical sensors, particularly those based on ion transfer at the interface between two immiscible electrolyte solutions (ITIES), offer significant advantages such as high selectivity, ease of fabrication, and cost effectiveness for toxic metal ion detection. However, distinguishing between cyclic voltammograms (CVs) of analytes with closely spaced half-wave potentials, such as Cd2+ and Cu2+, remains a challenge, especially for non-expert users. In this work, we present a novel methodology that integrates advanced artificial intelligence (AI) models with ITIES-based sensing to automate and enhance metal ion detection. Our approach first employed a convolutional neural network to classify CVs as either ideal or faulty with an accuracy exceeding 95 percent. Ideal CVs were then further analyzed for metal ion identification, achieving a classification accuracy of 99.15 percent between Cd2+ and Cu2+ responses. Following classification, an artificial neural network was used to quantitatively predict metal ion concentrations, yielding low mean absolute errors of 0.0158 for Cd2+ and 0.0127 for Cu2+. This integrated AI–ITIES system not only provides a scientific methodology for differentiating analyte responses based on electrochemical signatures but also substantially lowers the expertise barrier for sensor signal interpretation. To our knowledge, this is the first report of the AI-assisted differentiation and quantification of metal ions from ITIES-based CVs, establishing a robust framework for the future development of user-friendly, automated electrochemical sensing platforms for environmental and biological applications. Full article
Show Figures

Figure 1

20 pages, 682 KiB  
Review
Detecting Abnormal Behavior Events and Gatherings in Public Spaces Using Deep Learning: A Review
by Rafael Rodrigo-Guillen, Nahuel Garcia-D’Urso, Higinio Mora-Mora and Jorge Azorin-Lopez
J. Sens. Actuator Netw. 2025, 14(4), 69; https://doi.org/10.3390/jsan14040069 - 2 Jul 2025
Viewed by 1106
Abstract
Public security is a crucial aspect of maintaining social order. Although crime rates in Western cultures may be considered socially acceptable, it is important to continually improve security measures to prevent potential risks. With the advancements in artificial intelligence methods, particularly in deep [...] Read more.
Public security is a crucial aspect of maintaining social order. Although crime rates in Western cultures may be considered socially acceptable, it is important to continually improve security measures to prevent potential risks. With the advancements in artificial intelligence methods, particularly in deep learning and computer vision, it has become possible to detect abnormal event patterns in groups of people. This paper presents a review of the deep learning techniques employed for identifying gatherings of people and detecting anomalous events to enhance public security. Some of the open research areas are identified, including the lack of works addressing multiple cases of anomalies in large concentrations of people, which leaves open an important avenue for future scientific work. Full article
Show Figures

Figure 1

18 pages, 1572 KiB  
Case Report
Graphene–PLA Printed Sensor Combined with XR and the IoT for Enhanced Temperature Monitoring: A Case Study
by Rohith J. Krishnamurthy and Abbas S. Milani
J. Sens. Actuator Netw. 2025, 14(4), 68; https://doi.org/10.3390/jsan14040068 - 30 Jun 2025
Viewed by 593
Abstract
This case study aims to combine the advantage of the additive manufacturing of sensors with a mixed reality (MR) app, developed in a lab-scale workshop, to safely monitor and control the temperature of parts. Namely, the measurements were carried out in real time [...] Read more.
This case study aims to combine the advantage of the additive manufacturing of sensors with a mixed reality (MR) app, developed in a lab-scale workshop, to safely monitor and control the temperature of parts. Namely, the measurements were carried out in real time via a 3D-printed graphene–PLA nanocomposite sensor and communicated wirelessly using a low-power microcontroller with the IoT capability, and then transferred to the user display in the MR. In order to investigate the performance of the proposed computer-mediated reality, a user experience experiment (n = 8) was conducted. Statistical analysis results showed that the system leads to faster (>2.2 times) and more accurate (>82%) temperature control and monitoring by the users, as compared to the conventional technique using a thermal camera. Using a Holistic Presence Questionnaire (HPQ) scale, the users’ experience/training was significantly improved, while they reported less fatigue by 50%. Full article
(This article belongs to the Section Actuators, Sensors and Devices)
Show Figures

Figure 1

23 pages, 1560 KiB  
Article
Practical Aspects of Cross-Vendor TSN Time Synchronization Using IEEE 802.1AS
by Kilian Brunner, Florian Frick, Martin Ostertag and Armin Lechler
J. Sens. Actuator Netw. 2025, 14(4), 67; https://doi.org/10.3390/jsan14040067 - 30 Jun 2025
Viewed by 945
Abstract
Multi-vendor interoperability is essential for the stable operation, scalability, and successful market adoption of Time-Sensitive Networking (TSN). Conformance tests address protocol conformance. Informal interoperability testing and plugfests help to improve the quality and interoperability of specific implementations, and of the underlying international standard [...] Read more.
Multi-vendor interoperability is essential for the stable operation, scalability, and successful market adoption of Time-Sensitive Networking (TSN). Conformance tests address protocol conformance. Informal interoperability testing and plugfests help to improve the quality and interoperability of specific implementations, and of the underlying international standard documents. This paper presents three findings related to time synchronization in a multi-vendor TSN system. Differing interpretations of released standards and inconsistent setting of relevant system parameters resulted in undesirable behavior impacting the performance of the complete TSN system. The findings relevant to the standards themselves have been submitted to IEEE as maintenance items or are already being considered in work in progress at IEEE. In addition to interoperability testing, the importance of consistent system engineering and industry-specific TSN profiles are identified as important ingredients for successful implementation of TSN-based systems. Full article
(This article belongs to the Section Communications and Networking)
Show Figures

Figure 1

25 pages, 6794 KiB  
Article
Animal-Borne Adaptive Acoustic Monitoring
by Devin Jean, Jesse Turner, Will Hedgecock, György Kalmár, George Wittemyer and Ákos Lédeczi
J. Sens. Actuator Netw. 2025, 14(4), 66; https://doi.org/10.3390/jsan14040066 - 24 Jun 2025
Viewed by 974
Abstract
Animal-borne acoustic sensors provide valuable insights into wildlife behavior and environments but face significant power and storage constraints that limit deployment duration. We present a novel adaptive acoustic monitoring system designed for long-term, real-time observation of wildlife. Our approach combines low-power hardware, configurable [...] Read more.
Animal-borne acoustic sensors provide valuable insights into wildlife behavior and environments but face significant power and storage constraints that limit deployment duration. We present a novel adaptive acoustic monitoring system designed for long-term, real-time observation of wildlife. Our approach combines low-power hardware, configurable firmware, and an unsupervised machine learning algorithm that intelligently filters acoustic data to prioritize novel or rare sounds while reducing redundant storage. The system employs a variational autoencoder to project audio features into a low-dimensional space, followed by adaptive clustering to identify events of interest. Simulation results demonstrate the system’s ability to normalize the collection of acoustic events across varying abundance levels, with rare events retained at rates of 80–85% while frequent sounds are reduced to 3–10% retention. Initial field deployments on caribou, African elephants, and bighorn sheep show promising application across diverse species and ecological contexts. Power consumption analysis indicates the need for additional optimization to achieve multi-month deployments. This technology enables the creation of novel wilderness datasets while addressing the limitations of traditional static acoustic monitoring approaches, offering new possibilities for wildlife research, ecosystem monitoring, and conservation efforts. Full article
Show Figures

Figure 1

27 pages, 611 KiB  
Review
Edge Computing and Its Application in Robotics: A Survey
by Nazish Tahir and Ramviyas Parasuraman
J. Sens. Actuator Netw. 2025, 14(4), 65; https://doi.org/10.3390/jsan14040065 - 23 Jun 2025
Viewed by 1082
Abstract
The edge computing paradigm has gained prominence in both academic and industry circles in recent years. When edge computing facilities and services are implemented in robotics, they become a key enabler in the deployment of artificial intelligence applications to robots. Time-sensitive robotics applications [...] Read more.
The edge computing paradigm has gained prominence in both academic and industry circles in recent years. When edge computing facilities and services are implemented in robotics, they become a key enabler in the deployment of artificial intelligence applications to robots. Time-sensitive robotics applications benefit from the reduced latency, mobility, and location awareness provided by the edge computing paradigm, which enables real-time data processing and intelligence at the network’s edge. While the advantages of integrating edge computing into robotics are numerous, there has been no recent survey that comprehensively examines these benefits. This paper aims to bridge that gap by highlighting important work in the domain of edge robotics, examining recent advancements, and offering deeper insight into the challenges and motivations behind both current and emerging solutions. In particular, this article provides a comprehensive evaluation of recent developments in edge robotics, with an emphasis on fundamental applications, providing in-depth analysis of the key motivations, challenges, and future directions in this rapidly evolving domain. It also explores the importance of edge computing in real-world robotics scenarios where rapid response times are critical. Finally, the paper outlines various open research challenges in the field of edge robotics. Full article
Show Figures

Figure 1

17 pages, 4425 KiB  
Article
Design and Implementation of a Secure Communication Architecture for IoT Devices
by Cezar-Gabriel Dumitrache and Petre Anghelescu
J. Sens. Actuator Netw. 2025, 14(4), 64; https://doi.org/10.3390/jsan14040064 - 23 Jun 2025
Viewed by 635
Abstract
This paper explores the integration of Internet of Things (IoT) devices into modern cybersecurity frameworks, and it is intended to be a binder for the incorporation of these devices into emerging cybersecurity paradigms. Most IoT devices rely on WPA2-personal protocol, a wireless protocol [...] Read more.
This paper explores the integration of Internet of Things (IoT) devices into modern cybersecurity frameworks, and it is intended to be a binder for the incorporation of these devices into emerging cybersecurity paradigms. Most IoT devices rely on WPA2-personal protocol, a wireless protocol with known security flaws, being effortless to penetrate by using various specific tools. Through this paper, we proposed the use of two Raspberry Pi platforms, with the help of which we created a secure wireless connection by implementing the 802.1X protocol and using digital certificates. Implementing this type of architecture and the devices used, we obtained huge benefits from the point of view of security and energy consumption. We tested multiple authentication methods, including EAP-TLS and EAP-MSCHAPv2, with the Raspberry Pi acting as an authentication server and certificate manager. Performance metrics such as power consumption, latency, and network throughput were analysed, confirming the architecture’s effectiveness and scalability for larger IoT deployments. Full article
Show Figures

Figure 1

Previous Issue
Back to TopTop