Search Results (11,747)

IoT devices operate as integrated systems spanning hardware, firmware/software layers, and communication layers. In operational settings, many faults and performance degradations are emergent: they arise from cross-layer interactions, workload changes, and telemetry artifacts, rather than a single physics-of-failure mechanism. These realities make traditional supervised fault classification difficult because labeled fault data are rarely available during deployment, and the fault surface is unknown and a priori. This paper presents a practitioner-oriented, label-free fault detection and diagnosis (FDD) pattern based on Dynamic Time Warping (DTW) for rapid implementation in production IoT telemetry. The method represents a device as a sequence of overlapping episodes and organizes telemetry into interpretable layers (hardware sensors, communication health proxies, and software/firmware-derived KPIs). A reference library of regular episodes is built from an assumed-healthy training window; new episodes are scored using constrained DTW distances against this library, while retaining per-layer and per-channel contributions for attribution. We show that production performance depends strongly on operational parameterization, including episode length, DTW constraints, robust threshold learning, and temporal validation. Within a verified-healthy evaluation window, the tuned configuration achieves an AUROC of 0.97 for the temporally structured faults DTW is suited to (bias, drift, and interaction faults, with spikes detected at an AUROC of 0.93), detecting 100% of injected faults, with a mean delay under 25 min. We further show that constant-value (stuck-at) and missing-data (dropout) faults fall outside DTW’s shape-matching scope (AUROC about 0.66) and are better served by complementary variance- and missingness-based detectors, a consequence of DTW’s shape-matching scope rather than a parameter choice. This work contributes a system-level methodological framework for deploying DTW as an IoT fault-detection-and-diagnosis capability: an episode-and-layer architecture aligned with hardware, communication, and software/firmware ownership; a label-free reference library requiring only assumed-healthy data; per-layer and per-channel attribution for cross-domain triage; and a reproducible operational tuning procedure. Together, these deliver a fast-to-deploy, scalable, and accurate first-line detector for label-scarce IoT systems. Full article

With the widespread adoption of intelligent terminals, edge devices, and distributed information systems in the financial domain, financial security sensing data exhibit multisource heterogeneity, dynamic temporal patterns, and high privacy sensitivity. Traditional centralized anomaly detection methods are no longer able to simultaneously satisfy the requirements of cross-institutional or cross-node collaborative modeling, client data privacy protection, and robust monitoring of transaction and system anomalies. To address this challenge, a data-local federated deep anomaly detection framework has been proposed for distributed financial security sensing systems. Initially, a local deep financial security sensing representation module is constructed to perform temporal encoding and attention-based modeling on multisource financial signals, including terminal operation status, network transaction communication, backend server operation, identity authentication, and anomaly alerts, thereby extracting representations relevant to anomalous behaviors. Subsequently, a data-local federated optimization and personalized aggregation mechanism is developed to enable cross-node knowledge sharing without transmitting raw transaction or client data, while local personalized detection heads are employed to adapt to non-independent and identically distributed (non-IID) financial institution data. Furthermore, an adversarially robust security detection and trust-aware aggregation strategy is introduced to enhance model stability under input noise, feature masking, anomaly camouflage, and potential malicious client updates. Experimental results demonstrate that the proposed method achieves an Accuracy of $92.37\%$, a Precision of $89.41\%$, a Recall of $88.26\%$, an F1-score of $88.83\%$, an AUC of $93.06\%$, and a PR-AUC of $89.15\%$ in the primary financial anomaly detection task, significantly outperforming baseline methods such as Isolation Forest, Autoencoder, LSTM, Transformer, FedAvg, FedProx, SCAFFOLD, and MOON. In robustness experiments, the method attains F1-scores of $87.95\%$, $86.42\%$, $86.88\%$, $84.57\%$, $86.73\%$, and $83.91\%$ under Gaussian noise, feature masking, temporal shift, adversarial perturbation, and $20\%$ and $30\%$ malicious client scenarios, respectively. Ablation studies further confirm the effectiveness of local representation learning, personalized federated optimization, adversarial training, and trust-aware aggregation mechanisms. Overall, the proposed approach provides an efficient intelligent anomaly detection solution for financial AI security monitoring scenarios characterized by data localization requirements, node heterogeneity, and attack perturbations. Full article

Technological advances in hardware and software have enabled the development of novel in situ plankton imaging systems to investigate the spatial and temporal distribution of plankton communities. State-of-the-art machine learning approaches have been applied for automated image classification, effectively handling the complex and highly variable morphology of plankton while maintaining high accuracy. Despite these advances, few instruments can acquire zooplankton images autonomously in a continuous underway mode, which is essential for large-scale oceanographic surveys conducted aboard research vessels or ships of opportunity. Here, we present SiMFlux, an underway shadowgraphic imaging system developed at the University of São Paulo, and report results from the Orient Expedition. Observations were conducted aboard an 80-foot sailing vessel navigating across the Indian and Atlantic Oceans. A total of 193 videos were analyzed from daily route segments, yielding over 1.2 million regions of interest (ROIs) containing organisms and detrital particles. Particles were automatically classified and subsequently validated by plankton experts. Full article

This study examines how German textbooks provide learning-behavioral affordances for sustainability-oriented intercultural competence development. Drawing on Klieme’s competence-model logic, ESD, intercultural competence research, learning behavior theory, and affordance theory, it treats “sustainable intercultural competence” not as a standardized construct but as a working shorthand for the sustainability-oriented development of intercultural competence. Methodologically, the study adopts a directed qualitative content analysis supplemented by descriptive frequency aggregation. All 37 units across the four volumes of Meilenstein were coded on a 0–2 scale across three affordance dimensions: cognitive-understanding affordance, reflective value-judgment affordance, and interaction-action affordance. The findings show that the series provides substantial but uneven affordances. Interaction-action received the highest aggregated score, followed by cognitive-understanding, whereas reflective value-judgment remained substantially lower. Units on family, identity, sustainability, and civic engagement offer the most balanced affordance structures, whereas everyday practical units privilege communicative action and disciplinary units privilege cognitive understanding. The study argues that textbook-based intercultural learning should be examined not only through topic inclusion but also through how texts, prompts, and tasks organize opportunities for comparison, reflection, judgment, negotiation, and action. Full article

Accurate rainfall monitoring is vital for hydrology and environmental sensing. This study presents a physics-aware deep learning framework using E-band (71–86 GHz) commercial microwave links (CMLs). Using the extensive urban CML dataset and methodology, a bi-directional Long Short-Term Memory (Bi-LSTM) model is developed to classify wet and dry periods under a temporal generalization framework across heterogeneous link configurations. The approach integrates physical signal decomposition, including baseline estimation, gaseous attenuation correction, and wet antenna attenuation (WAA) modeling, with sequence-based learning. Results demonstrate that the temporal deep learning model outperforms classical threshold-based and physical k–R approaches when evaluated over independent temporal validation blocks, effectively reducing sensitivity to path-length-related variability on heterogeneous paths. The model maintains stable performance (loss < 3%) under moderate signal-level noise. SHapley Additive exPlanations (SHAP) confirm the model relies on physical features, such as signal volatility and temporal trends, to reliably differentiate rainfall from WAA. This framework highlights the potential of E-band infrastructure as a distributed sensing network for integrated sensing and communication (ISAC) architectures. Full article

This paper investigates the output synchronization of fractional-order cyber-physical systems (FOCPSs) under communication constraints. To address limited bandwidth and high transmission costs, an event-triggered encoding-decoding sampled-data iterative learning control (ET-EDSDILC) protocol is proposed. The control law integrates a quantized sampling framework with an encoding–decoding mechanism to reconstruct control signals and address communication constraints. Furthermore, an event-triggered mechanism based on error energy attenuation (EEA) is developed to adjust communication frequency by monitoring error trends, thereby reducing unnecessary data transmissions. By applying fractional-order calculus and the contraction mapping principle, sufficient conditions for output synchronization are derived. Numerical simulations show that the proposed ET-EDSDILC framework reduces communication overhead and data redundancy while maintaining tracking performance, offering a solution for FOCPSs under communication constraints. Full article

Resource allocation for cognitive Vehicle-to-Everything (V2X) networks is challenging due to dynamic spectrum sharing, strong interference coupling, and stringent latency constraints for safety-critical Vehicle-to-Vehicle (V2V) traffic. Although recent Multi-Agent Reinforcement Learning (MARL) approaches report promising gains, many evaluations are conducted at limited and fixed network scales, which restricts insights into scalability under dense spectrum reuse. This paper investigates cooperative multi-agent learning for interference-aware and deadline-constrained V2X resource management. We propose a Q-value Transformation (QTRAN)-based value decomposition framework under centralized training with decentralized execution (CTDE) for joint resource-block and power allocation among V2V agents. The proposed approach is implemented in a realistic V2V/V2I simulator incorporating Manhattan grid mobility, fast fading, explicit cross-tier and co-channel interference, and per-link payload/deadline dynamics. Beyond communication-level performance, improved timely delivery of V2V safety messages can support cooperative maneuvering, collision avoidance, platooning, and infrastructure-assisted traffic management. Extensive simulations across varying numbers of V2V agents benchmark QTRAN against independent learning baselines including MARL and centralized single-agent learning (SARL). Results show that QTRAN improves performance compared with the selected learning baselines and enhances the throughput–reliability trade-off under interference-coupled spectrum reuse. For instance, at $N_{\mathrm{V}2\mathrm{V}}=20$, QTRAN achieves a V2V rate of $0.194\pm 0.004$ and a V2I rate of $9.117\pm 0.213$, while reaching a V2V success rate of $0.812\pm 0.017$ with a low Deadline Miss Ratio of $0.001\pm 0.000$. At higher density ($N_{\mathrm{V}2\mathrm{V}}=50$), QTRAN sustains strong reliability (V2V success rate of $0.719\pm 0.006$ and Completion Ratio of $0.716\pm 0.006$) while maintaining competitive infrastructure throughput (V2I rate of $9.251\pm 0.114$). These results indicate that QTRAN effectively captures non-linear interference interactions, enabling coordinated decentralized spectrum and power decisions under the adopted density-based evaluation setting, thereby enhancing V2V reliability and throughput in cognitive Internet of Vehicles. Full article

Pharmaceutical cold-chain distribution must maintain timely access to temperature-sensitive medicines while limiting the energy demand and carbon emissions associated with refrigerated transport. This study proposes a sustainable dynamic route optimization method that integrates reinforcement learning (RL) with an improved neighborhood search (NS) algorithm to balance delivery timeliness and transportation carbon emissions. The NS algorithm is enhanced with carbon emission and timeliness operators, and RL adaptively adjusts their weights under dynamic events, including traffic congestion, vehicle failure, and order insertion. The method is evaluated using the Solomon Benchmark dataset and a warehouse-to-community-pharmacy last-mile distribution case for chronic-disease medicines. The RL-NS algorithm achieves an average computation time of 45.3 ms and a standard deviation of 2.7, outperforming the comparison algorithms. In the case study, it reduces transportation carbon emissions by approximately 18% and delivery time by approximately 12% relative to traditional routing. By reducing route redundancy and enabling rapid replanning, the method supports lower-emission and potentially more energy-efficient transport operations. The findings demonstrate its relevance to sustainable transportation, sustainable logistics, and resilient pharmaceutical cold-chain management. Full article

The proliferation of Unmanned Aerial Vehicles (UAVs) in various applications has created a pressing need for robust and efficient communication systems. Fifth-generation (5G) networks can support UAV connectivity through high bandwidth and low-latency communication; however, rapid three-dimensional UAV mobility creates handover-management challenges that can increase signalling overhead, service interruption, and Quality of Service (QoS) degradation. This paper presents an integrated framework that combines LSTM-based multi-UAV trajectory prediction with proactive handover optimization using an Advantage Actor–Critic (A2C) Deep Reinforcement Learning (DRL) agent. The LSTM predictor is evaluated on a real-world UAV trajectory dataset and reports a root mean square error (RMSE) of 4.37 m over a 5 s prediction horizon after conversion to a local East–North–Up coordinate frame. A lightweight simulation-level coordination mechanism is included to reduce simultaneous target-cell contention among multiple UAVs; it is not claimed as a new standardized 3GPP signalling procedure. Handover performance is evaluated by replaying 180 held-out flight trajectories in a controlled 5G simulation across ten independent random seeds. Under these stated assumptions, the proposed framework achieves a handover success rate of $94.2\pm 0.8$%, an average SINR of $15.8\pm 0.2$ dB, a handover delay of $45.2\pm 1.1$ ms, and a handover frequency of $0.85\pm 0.05$ HOs/min, outperforming the tuned 3GPP A3, reactive SINR, and CASH baselines in the reported simulation results (Wilcoxon signed-rank test, $p<0.01$, Bonferroni-corrected). The experimental setup is described in detail to support methodological transparency and facilitate future replication, but the handover results should be interpreted as simulation-based evidence rather than live-network validation. Full article

Engineering education prepares graduates to face complex environmental and societal challenges. This involves the intersection of sustainability and innovation. Integrating these agendas is therefore necessary, and this involves identifying specific elements that have not yet been explored. To examine this relationship, a systematic literature review was conducted using an adapted PRISMA 2020 approach appropriate for a bibliometric and thematic systematic review, through four research questions related to knowledge production, pedagogical methods, innovation outcomes, and reported results. The PRISMA phases were adopted using the SCOPUS and ERIC databases. This yielded three clusters: innovation, sustainability, and engineering education. Student-centered pedagogies have also been identified as an explored opportunity to enhance innovation skills aligned with sustainability objectives. However, this incorporation involves many elements to explore, including the connection between innovation outcomes and sustainability impact. This context involves both development and the relationships among individuals, institutions, and ecosystems. This requires managing diverse visions, languages, and cultures, which highlights several challenges: long-term impacts, mindset development, contextual influences, pedagogical strategies, research–practice alignment, stakeholder communication, and faculty preparation. Overall, the findings show progress but reveal challenges across approaches and contexts. This is because sustainability-driven innovation in engineering education requires coordinated curricular, institutional, and ecosystem-oriented strategies to support learning and strengthen contributions to sustainable futures. Full article

This AI-assisted integrative literature review, grounded in complexity theory, examines how existing scholarship conceptualizes the potential of scenario planning to support future-ready mindsets, shared language, symbolic anchors, and adaptive capacity in educational leadership. Using the Consensus AI-assisted research synthesis platform, peer-reviewed literature was identified, ranked for semantic relevance, and screened in relation to three guiding questions focused on scenario planning, VUCA leadership, shared language, and professional learning. From an initial corpus of 2092 papers, 100 high-relevance studies were purposively selected for full review and analyzed through narrative thematic synthesis. Findings suggest that scenario planning may contribute to foresight, adaptability, and communicative capacity when facilitated inclusively and used iteratively with an equity focus. When treated as an ongoing practice rather than a one-time planning exercise, scenario planning appears to offer a promising structure for cultivating adaptive, innovative leadership capable of navigating complex educational change. Full article

Wearable technologies offer practical opportunities for assistive communication and educational support in introductory sign language learning. This paper presents an IoT-enabled smart glove for deterministic static sign language recognition over a bounded vocabulary of 15 isolated static gestures, comprising digits (0–9) and five vowel handshapes (A, E, I, O, U). The system is intended for foundational static gesture and posture practice and is not designed or validated for dynamic gestures, coarticulated signing, continuous sign language recognition, or sentence-level translation. The prototype integrates five 2.2-inch (55.9 mm) resistive flex sensors and an MPU6050 3-axis accelerometer, performs acquisition, exponential moving average filtering, user-specific calibration, normalization, and deterministic classification on a NodeMCU ESP32 board, and transmits selected processed variables to Arduino Cloud through MQTT for remote monitoring. A 10 s calibration routine maps user-specific open-hand and closed-fist responses into normalized flex-sensor ranges, allowing the same deterministic rule structure to operate across participants without model retraining. Experimental evaluation with 10 healthy adult participants aged 20–41 years (mean age: 27 years), all familiar with sign language and all providing written informed consent, produced a balanced dataset of 1500 labeled steady-state sensor vectors. The class-averaged recognition rate was 92.8%, and leave-one-subject-out validation produced a subject-wise accuracy of $92.80\pm 2.03\%$, with individual participant accuracies ranging from 90.00% to 96.00%. The local embedded processing pipeline required less than 2 ms per cycle, the complete path including MQTT visualization produced approximately 150 ms end-to-end latency, and the device operated for up to 14 h using a 3.7 V, 1000 mAh Li-Po battery. The results indicate that calibrated deterministic sensor fusion can provide a low-cost, low-latency, edge-executed solution for bounded static sign-language gesture learning tasks while maintaining stable short-term subject-wise performance under controlled experimental conditions. Full article

Contemporary youth mental health challenges necessitate collaborative approaches to prevention and support. The Mental Health Ambassador (MHA) program was developed to equip youth and adults in mentoring programs with the skills and confidence to discuss and support youth mental health, with the broader goal of creating meaningful impact within their local communities. The 8-session MHA program involves group-based learning and a local youth-led advocacy project. This mixed-method pre-post pilot evaluation examined program acceptability, implementation experiences, and potential for effectiveness. Participants included youth aged 14–18 years (n = 9) and adult mentors (n = 11) from mentoring organizations across five counties in Minnesota. Quantitative surveys assessed mental health resource awareness, preparedness to address youth mental health concerns, confidence in engaging in mental health conversations, and confidence in providing resources and referrals. Post-program focus groups explored participants’ experiences, perceived benefits, and implementation challenges. Findings indicated that both youth and adult participants reported positive experiences with the program and demonstrated increases in resource awareness, preparedness to address youth mental health concerns, confidence in discussing mental health, and confidence in providing resources and referrals. Qualitative findings further highlighted the value of youth-led advocacy activities and identified key considerations for implementation within mentoring settings. Mentoring programs may represent an ideal context for equipping youth and adults to provide early, community-based support for youth mental health concerns. Full article

This study investigates the use of interpretable machine learning techniques to identify behavioral and demographic patterns associated with diabetes, based on structured population survey data from the Canadian Community Health Survey (CCHS). A decision tree classifier was applied to a dataset comprising $16,824$ respondents and 38 preprocessed features covering lifestyle, well-being, and sociodemographic factors. The model was optimized through grid search with five-fold stratified cross-validation, achieving a test accuracy of $61.3$% (mean $62.6$% ±$0.6$% across a $10\times 5$ repeated stratified cross-validation). Feature importance analysis revealed that age, alcohol consumption patterns, daily energy expenditure, and physical activity were the most influential factors associated with diabetes status, with the top three features exhibiting stable importance across all cross-validation folds. The model produced a set of 32 human-readable decision rules; a sensitivity analysis confirmed that these rules are stable across encoding choices and cross-validation folds. Several model variants were evaluated: a class-weighted decision tree, a logistic regression baseline, an age-only decision tree, and an age and sex logistic regression. The class-weighted model improved minority-class recall (from $0.25$ to $0.53$) at the cost of overall accuracy. A one-hot encoding sensitivity analysis showed that replacing ordinal label encoding of nominal variables with one-hot encoding produces virtually identical results (accuracy: $61.4$% vs. $61.3$%), confirming that the main rules are not artifacts of the encoding choice. Although the classification accuracy is moderate and not significantly better than a majority-class baseline (McNemar’s test, $p=0.455$), the extracted rules confirmed several known associations and revealed interactions between social and lifestyle variables. These rules are intended as hypothesis-generating population-level descriptors rather than validated clinical decision tools, and no causal inference is claimed. This approach demonstrates the value of rule-based models for exploratory public health research. Full article

Optical modules serve as the core transmission interfaces for artificial intelligence computing networks and digital communications. In recent years, demand for these modules has experienced explosive growth. During mass production, the requirements for the accuracy of surface defect detection and noise resistance have continued to rise. Existing POL detection models are susceptible to environmental noise interference; effective defect information is easily overwhelmed by noise entropy, and these models exhibit a high false negative rate for low-contrast and minute defects. This paper proposes a traditional image processing detection scheme that incorporates information entropy constraints. All experimental samples were collected from actual industrial mass production lines. The core process includes: noise suppression during the calibration stage using an entropy-weighted Hough transform; Canny edge detection combined with local entropy filtering for contour localization; and defect fusion recognition based on Hu similarity matching and entropy difference verification. Experimental results show that, compared to traditional POL methods, the proposed approach (WOMC) achieves an average improvement of 35.77% in image clarity and approximately a 2.25-fold increase in detection rate under Gaussian and salt-and-pepper noise conditions. According to statistical analysis of the experiments, this method achieved an accuracy of 96.67%, a recall rate of 97.32%, and a false positive rate of 3.31% in defect detection. In addition, the comprehensive performance score of this detection model reached 96.99%. Moreover, it does not require the deployment of deep-learning models, has a low computing power cost, and is suitable for the detection requirements of large-scale mass production. Full article