Search Results (263)

Background/Objectives: The study primarily examined whether an IoT-based medication assistance system enhances ART adherence relative to standard care, and secondarily evaluated device feasibility and error patterns over time. Methods: This prospective study was conducted between June 2022 and October 2023 at a tertiary hospital in South Korea. Adults (≥19 years) living with HIV and prescribed ART were included; those with comorbid hepatitis B or C were excluded. People living with HIV who agreed to use the IoT-based InPHRPILL system (Sofnet Inc., Seoul, Republic of Korea) were assigned to the intervention group, whereas those who declined were assigned to the control group. Viral suppression, CD4⁺ cell counts, and adherence rates were measured. Additional analyses evaluated 12-month longitudinal adherence using pill-count data in both groups, and device-measured adherence and device-associated error rates in the intervention group. Results: Thirty-five participants (12 in the intervention group and 23 in the control group) were included. The intervention group demonstrated marginally shorter durations since HIV diagnosis and ART initiation at study enrollment, as well as slightly higher baseline HIV-RNA levels; however, these differences did not reach statistical significance. The median pill-counting and IoT device adherence rates were 100% and 87.4%, respectively (median deviation error rate = 4.4%). Poisson regression revealed significantly reduced error rates over time (β = −0.06493, p < 0.01), suggesting improved device use proficiency. Conclusions: IoT-based medication assistance systems may provide objective, real-time monitoring of ART adherence and facilitate identification of discrepancies between clinical evaluations and actual adherence patterns. Larger studies targeting individuals with suboptimal adherence are warranted to determine whether such systems can enhance adherence outcomes. Full article

Hazardous gas detection requires portable, low-power sensors with high sensitivity, where micro-heater design is critical for semiconductor metal oxide (SMO) sensors. This study presents a standardized evaluation framework for quantitatively comparing patterned micro-heaters under equal-power conditions, ensuring objective comparison across geometries. Two key metrics—power efficiency and temperature uniformity—were defined, normalized, and integrated into a single optimal score through weighted summation. The framework was validated through coupled electro-thermal simulations and experiments on six geometries, including spiral and meander patterns. Results demonstrated that the framework enables accurate identification of designs combining low power consumption with high temperature uniformity. Notably, the meander-based design showed superior efficiency and uniformity, demonstrating its suitability for practical applications. This framework thus offers a rational tool for micro-heater design, supporting the development of reliable, energy-efficient devices for portable and Internet of Things (IoT) applications. Full article

With the accelerated expansion of the Internet of Things (IoT), massive distributed and heterogeneous devices are increasingly exposed to severe security threats. Traditional centralized intrusion detection systems (IDS) suffer from significant limitations in terms of privacy preservation and communication overhead. Federated Learning (FL) offers an effective paradigm for building the next generation of distributed IDS; however, it remains vulnerable to poisoning attacks in open environments, and existing client selection strategies generally lack robustness and security awareness. To address these challenges, this paper proposes an Uncertainty-Guided Evolutionary Game-Theoretic (UEGT) Client Selection mechanism. Built upon evolutionary game theory, UEGT integrates Shapley value, gradient similarity, and data quality to construct a multidimensional payoff function and employs a replicator dynamics mechanism to adaptively optimize client participation probabilities. Furthermore, uncertainty modeling is introduced to enhance strategic exploration and improve the identification accuracy of potentially high-value clients. Experimental results under adversarial scenarios demonstrate that UEGT maintains stable convergence even under a high fraction of malicious participating clients, achieving an average accuracy exceeding 89%, which outperforms several mainstream client selection and robust aggregation methods. Full article

IoT applications are increasingly common, yet they often rely on expensive, externally managed authentication services. This paper introduces a novel, self-contained authentication method for IoT applications which leverages fog computing principles to lower operational costs and infrastructure complexity. The proposed system, fogauth, combines device serial numbers with cryptographically generated UUIDs to establish secure identification without third-party services. A static cloud-side architecture coupled with a lightweight, locally hosted API enables secure authentication through object-storage operations. Performance testing demonstrates comparable security performance to commercial cloud-based authentication while reducing long-term operational costs and maintaining latency at below 2 minutes in production conditions. fogauth provides a scalable and economically viable alternative for companies seeking to reduce cloud dependency and minimize long-term costs associated with IoT application security. To support reproducibility, a complete open-source implementation and validation dataset are provided, allowing independent replication and extension of the system. Full article

The Internet of Things (IoT) continues to expand across critical infrastructures, enabling automation, efficiency, and data driven decision making; yet, reliable device identification from network traffic remains hampered by severe class imbalance that skews learning and degrades performance. Synthetic data generation offers a promising remedy, particularly in privacy-sensitive security settings where access to representative traffic is limited. This paper advances the state of the art by proposing a framework that unites graph-conditioned generative modeling with large language model (LLM) guidance to produce realistic, semantically valid synthetic network traffic for imbalanced classification. First, we construct feature relationship graphs derived from Pearson correlation, Spearman rank correlation, and mutual information to capture inter-feature dependencies, and use these graphs to condition a Wasserstein GAN (WGAN), thereby preserving structural properties of real traffic during generation. Second, we employ an LLM to define class-specific semantic constraints, including admissible feature ranges, attribute correlations, and protocol level rules, which are enforced as soft guidance to steer the generator toward label-consistent and standards-compliant samples. Third, we institute a dual validation loop that combines LLM-based feedback on constraint satisfaction with evaluation of classifiers trained on datasets balanced by our method versus the traditional SMOTE technique. Lastly, extensive experiments demonstrate that jointly leveraging structural (graph) and semantic (LLM) conditioning yields higher-fidelity synthetic traffic and delivers consistent gains in macro-F1 and balanced accuracy for network traffic classification, highlighting the framework’s utility for security analytics under data scarcity and privacy constraints. Full article

Industry 5.0 represents a paradigm shift toward human–AI collaboration in manufacturing, incorporating unprecedented volumes of robots, Internet of Things (IoT) devices, Augmented/Virtual Reality (AR/VR) systems, and smart devices. This extensive interconnectivity introduces significant cybersecurity vulnerabilities. While AI has proven effective for cybersecurity applications, including intrusion detection, malware identification, and phishing prevention, cybersecurity professionals have shown reluctance toward adopting black-box machine learning solutions due to their opacity. This hesitation has accelerated the development of explainable artificial intelligence (XAI) techniques that provide transparency into AI decision-making processes. This systematic review examines XAI-based intrusion detection systems (IDSs) for Industry 5.0 environments. We analyze how explainability impacts cybersecurity through the critical lens of adversarial XAI (Adv-XIDS) approaches. Our comprehensive analysis of 135 studies investigates XAI’s influence on both advanced deep learning and traditional shallow architectures for intrusion detection. We identify key challenges, opportunities, and research directions for implementing trustworthy XAI-based cybersecurity solutions in high-stakes Industry 5.0 applications. This rigorous analysis establishes a foundational framework to guide future research in this rapidly evolving domain. Full article

This research is dedicated to the development of an intelligent biometric system based on the synergy of Internet of Things (IoT) technologies and Artificial Intelligence (AI). The primary goal of this research is to explore the possibilities of personal identification using two distinct biometric traits: facial images and the venous pattern of the palm. These methods are treated as independent approaches, each relying on unique anatomical features of the human body. This study analyzes state-of-the-art methods in computer vision and neural network architectures and presents experimental results related to the extraction and comparison of biometric features. For each biometric modality, specific approaches to data collection, preprocessing, and analysis are proposed. We frame optimization in practical terms: selecting an edge-suitable backbone (ResNet-50) and employing metric learning (Triplet Loss) to improve convergence and generalization while adapting the stack for edge IoT deployment (Dockerized FastAPI with JWT). This clarifies that “optimization” in our title refers to model selection, loss design, and deployment efficiency on constrained devices. Additionally, the system’s architectural principles are described, including the design of the web interface and server infrastructure. The proposed solution demonstrates the potential of intelligent biometric technologies in applications such as automated access control systems, educational institutions, smart buildings, and other areas where high reliability and resistance to spoofing are essential. Full article

The field deployment of low-cost air quality sensor systems enables enhanced spatial resolution in air quality monitoring. Although these sensor systems cannot achieve the same accuracy as regulatory monitoring stations, they can attain acceptable levels of confidence and provide Indicative Measurements as regulated by Ambient Air Quality EU Directive. The integration of an anemometer into a system can provide additional information for the classification of the measurement area, the identification of potential sources of pollutant emissions, and the assessment of the device’s operating conditions during measurement. In this study, the measurement capabilities of an Airbox, a low-cost air quality sensor system, were extended through the integration of a DW6410 anemometer (Davis Instruments). The Airbox, designed to transmit data in real-time or near real-time to servers and IoT platforms, was deployed for a duration of 4 months, from October 2021 to February 2022, within the airport area of Grottaglie (Southern Italy). The anemometric measurements and particulate concentration data (PM_2.5 and PM₁₀, measured by NextPM sensor, Tera Sensor) were integrated and compared to meteorological open data and data from a regulatory regional air quality control network located in the area around the airport. Full article

Accurate and continuous tracking of athletes is essential to meet the infotainment demands and health and safety requirements of major marathon events. However, the current ability to track individual athletes or groups at mass sporting events is severely limited by the weight, size and cost of the equipment required. In marathons, Radio Frequency Identification (RFID) technology is typically used for timing but can only provide accurate tracking at widely spaced intervals, relying on heuristic and interpolation algorithms to estimate runners’ positions between measurement points. Alternative IOT solutions, such as Low Power Wide Area Network (LWPAN), have limitations in terms of range and require dedicated infrastructure and regulation. Therefore, we analyzed the potential use of smartwatches as accurate and continuous tracking devices for athletes, assessing battery consumption during tracking and standby drain, achievable GPS tracking accuracy and the update rate of data transfer from the device in urban environments. The 4G LTE battery drain is different from non-urban areas. Analysis of standby usage is necessary as devices need to conserve power for tracking. We programmed an application that allowed us to control the modalities of acquisition and transmission intervals, integrating advanced logging and statistics at runtime, and evaluated the achievable results in major marathon events. Our empirical evaluation at the Frankfurt, Athens and Vienna marathons with three different types of smartwatch tracking platforms showed the validity of this approach, while respecting some necessary limitations of the tracking settings. Median battery drain was 5.3%/h in standby before race start (σ 1.5) and 16.5%/h in tracking mode (σ 3.29), with an actual update rate varying between 19 and 57 s on Wear OS devices. The average GPS offset to the track was 4.5 m (σ 8.7). Future work will focus on integrating these consumer devices with existing time and tracking infrastructure. Full article

The implementation of the Internet of Things (IoT) in logistics has the ability to transform the whole logistics industry by improving business models, operational efficiency, traceability, security, and customer experience. The manual logistics process causing a lot of late deliveries, wrong deliveries, and line stoppages in a multinational automotive company. That led to the pursuit of this research work to convert the manual call-off process to a fully system-controlled process. The main objective of this research was to implement system-controlled warehouse call-offs and scheduling processes to reduce line stoppages caused by late and incorrect delivery of parts to the line, as well as hot call-offs, and to improve the overall efficiency of line supply routes. The introduction of IoT in the warehouse comes with a takted process, meaning that each step of the line supply process is timed. The process introduces scanners to support process confirmation and link every process step to System Applications and Products in Data Processing (SAP) to allow for traceability. The interconnected devices and system in this study connect line-side reality (using Rapid Frequency Identification (RFID), optic sensors, and the Integrated Production System Logistics (IPSL) bill of material information) with the SAP demand and part requirements. The IoT implementation results show a great improvement in the overall logistics of line supply processes. A decrease in line stoppages is witnessed, with a reduction of 69%, and line-side confirmation makes tracing easier, thereby enhancing process transparency. The addition of scanners provides line supply employees transparency with respect to where parts are going, further reducing the probability of wrong deliveries. Waste reduction is also a result of this research, as the takted processes allow for time saving on the round-trip time, which is reduced by 32%. Conclusively, this research adds to the expanding corpus of research on the application of IoT in logistics and offers useful advice to policymakers and logistics managers who wish to integrate IoT technologies into their operations. Full article

IoT (Internet of Things)-enabled solar irradiance sensors are evolving toward energy harvesting, interoperability, and open-source availability, yet current solutions remain either costly, closed, or limited in robustness. Based on a thorough literature review and identification of future trends, we propose an open-source smart wireless sensor that employs a small photovoltaic module simultaneously as sensing element and energy harvester. The device integrates an ESP32 microcontroller, precision ADC (Analog-to-Digital converter), and programmable load to sweep the PV (photovoltaic) I–V (Current–Voltage) curve and compute irradiance from electrical power and solar-cell temperature via a calibrated third-order polynomial. Supporting Modbus RTU (Remote Terminal Unit)/TCP (Transmission Control Protocol), MQTT (Message Queuing Telemetry Transport), and ZigBee, the sensor operates from batteries or supercapacitors through sleep–wake cycles. Validation against industrial irradiance meters across 0–1200 W/m² showed average errors below 5%, with deviations correlated to irradiance volatility and sampling cadence. All hardware, firmware, and data-processing tools are released as open source to enable reproducibility and distributed PV monitoring applications. Full article

With the quick development of network technology, the number of active IoT devices is growing rapidly. Numerous network scanning organizations have emerged to scan and detect network assets around the clock. This greatly facilitates illegal cyberattacks and adversely affects cybersecurity. Therefore, it is important to discover and identify network scanning organizations on the Internet. Motivated by this, we propose a network scanning organization discovery method based on a graph convolutional neural network, which can effectively cluster out network scanning organizations. First, we constructed a network scanning attribute graph to represent the topological relationship between network scanning behaviors and targets. Then, we extract the deep feature relationships in the attribute graph via graph convolutional neural network and perform clustering to get network scanning organizations. Finally, the effectiveness of the method proposed in this paper is experimentally verified with an accuracy of 83.41% for the identification of network scanning organizations. Full article

Molecular communication (MC) has emerged as a promising paradigm for nanoscale information exchange in Internet of Bio-Nano Things (IoBNT) environments, offering intrinsic biocompatibility and potential for real-time in vivo monitoring. This study proposes a cascaded MC channel framework for vascular stenosis detection, which integrates non-Newtonian blood rheology, bell-shaped constriction geometry, and adsorption–desorption dynamics. Path delay and path loss are introduced as quantitative metrics to characterize how structural narrowing and molecular interactions jointly affect signal propagation. On this basis, a peak response time-based delay inversion method is developed to estimate both the location and severity of stenosis. COMSOL 6.2 simulations demonstrate high spatial resolution and resilience to measurement noise across diverse vascular configurations. By linking nanoscale transport dynamics with system-level detection, the approach establishes a tractable pathway for the early identification of vascular anomalies. Beyond theoretical modeling, the framework underscores the translational potential of MC-based diagnostics. It provides a foundation for non-invasive vascular health monitoring in IoT-enabled biomedical systems with direct relevance to continuous screening and preventive cardiovascular care. Future in vitro and in vivo studies will be essential to validate feasibility and support integration with implantable or wearable biosensing devices, enabling real-time, personalized health management. Full article

Radio frequency (RF) fingerprinting, as an emerging physical layer security technology, demonstrates significant potential in the field of Internet of Things (IoT) security. However, most existing methods operate under a ‘closed-set’ assumption, failing to effectively address the continuous emergence of unknown devices in real-world scenarios. To tackle this challenge, this paper proposes an open-set radio frequency fingerprint identification (RFFI) method based on Multi-Task Prototype Learning (MTPL). The core of this method is a multi-task learning framework that simultaneously performs discriminative classification, generative reconstruction, and prototype clustering tasks through a deep network that integrates an encoder, a decoder, and a classifier. Specifically, the classification task aims to learn discriminative features with class separability, the generative reconstruction task aims to preserve intrinsic signal characteristics and enhance detection capability for out-of-distribution samples, and the prototype clustering task aims to promote compact intra-class distributions for known classes by minimizing the distance between samples and their class prototypes. This synergistic multi-task optimization mechanism effectively shapes a feature space highly conducive to open-set recognition. After training, instead of relying on direct classifier outputs, we propose to adopt extreme value theory (EVT) to statistically model the tail distribution of the minimum distances between known class samples and their prototypes, thereby adaptively determining a robust open-set discrimination threshold. Comprehensive experiments on a real-world dataset with 16 Wi-Fi devices show that the proposed method outperforms five mainstream open-set recognition methods, including SoftMax thresholding, OpenMax, and MLOSR, achieving a mean AUROC of 0.9918. This result is approximately 1.7 percentage points higher than the second-best method, demonstrating the effectiveness and superiority of the proposed approach for building secure and robust wireless authentication systems. This validates the effectiveness and superiority of our approach in building secure and robust wireless authentication systems. Full article

As IoT (internet of things) devices grow in prominence, safeguarding them from cyberattacks is becoming a pressing challenge. To bootstrap IoT security, device identification or authentication is crucial for establishing trusted connections among devices without prior trust. In this regard, radio frequency fingerprinting (RFF) is gaining attention because it is more efficient and requires fewer computational resources compared to resource-intensive cryptographic methods, such as digital signatures. RFF works by identifying unique manufacturing defects in the radio circuitry of IoT devices by analyzing over-the-air signals that embed these imperfections, allowing for the identification of the transmitting hardware. Recent studies on RFF often leverage advanced classification models, including classical machine learning techniques such as K-Nearest Neighbor (KNN) and Support Vector Machine (SVM), as well as modern deep learning architectures like Convolutional Neural Network (CNN). In particular, CNNs are well-suited as they use multidimensional mapping to detect and extract reliable fingerprints during the learning process. However, a significant limitation of these approaches is that they require large datasets and necessitate retraining when new devices not included in the initial training set are added. This retraining can cause service interruptions and is costly, especially in large-scale IoT networks. In this paper, we propose a novel solution to this problem: RFF using Siamese networks, which eliminates the need for retraining and allows for seamless authentication in IoT deployments. The proposed Siamese network is trained using in-phase and quadrature (I/Q) samples from 10 different Software-Defined Radios (SDRs). Additionally, we present a new algorithm, the Similarity-Based Embedding Classification (SBEC) for RFF. We present experimental results that demonstrate that the Siamese network effectively distinguishes between malicious and trusted devices with a remarkable 98% identification accuracy. Full article