Search Results (451)

Human Activity Recognition (HAR) has become crucial across various domains, including healthcare, smart homes, and security systems, owing to the proliferation of Internet of Things (IoT) devices. Several Machine Learning (ML) techniques, including Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM), have been proposed for HAR. However, they are still deficient in addressing the challenges of noisy features and insufficient data. This paper introduces a novel approach to tackle these two challenges, employing a Deep Learning (DL) Ensemble-Based Stacking Neural Network (SNN) combined with Generative Adversarial Networks (GANs) for HAR based on ambient sensors. Our proposed deep learning ensemble-based approach outperforms traditional ML techniques and enables robust and reliable recognition of activities in real-world scenarios. Comprehensive experiments conducted on six benchmark datasets from the CASAS smart home project demonstrate that the proposed stacking framework achieves superior accuracy on five out of six datasets when compared to literature-reported state-of-the-art baselines, with improvements ranging from 3.36 to 39.21 percentage points and an average gain of 13.28 percentage points. Although the baseline marginally outperforms the proposed models on one dataset (Aruba) in terms of accuracy, this exception does not alter the overall trend of consistent performance gains across diverse environments. Statistical significance of these improvements is further confirmed using the Wilcoxon signed-rank test. Moreover, the ASGAN-augmented models consistently improve macro-F1 performance over the corresponding baselines on five out of six datasets, while achieving comparable performance on the Milan dataset. The proposed GAN-based method further improves the activity recognition accuracy by a maximum of 4.77 percentage points, and an average of 1.28 percentage points compared to baseline models. By combining ensemble-based DL with GAN-generated synthetic data, a more robust and effective solution for ambient HAR addressing both accuracy and data imbalance challenges in real-world smart home settings is achieved. Full article

Applications of smart cities, smart buildings, smart agriculture systems, smart grids, and other smart systems benefit from Internet of Things (IoT) protocols, networks, and architecture. Wireless Sensor Networks (WSNs) in smart systems that employ IoT use wireless communication technologies between sensors in the Things layer and the Fog layer hub. Such wireless protocols and networks include WiFi, Bluetooth, and Zigbee, among others. However, the payload formats of these protocols are heterogeneous, and thus, they lack a unified frame format that ensures interoperability. In this paper, a lightweight, interoperable frame format for low-rate, small-size Wireless Sensor Networks (WSNs) in IoT-based systems is designed, implemented, and tested. The practicality of this system is underscored by the development of a gateway that transfers collected data from sensors that use the unified frame to online servers via message queuing and telemetry transport (MQTT) secured with transport layer security (TLS), ensuring interoperability using the JavaScript Object Notation (JSON) format. The proposed frame is tested using market-available technologies such as Bluetooth and Zigbee, and then applied to smart home applications. The smart home scenario is chosen because it encompasses various smart subsystems, such as healthcare monitoring systems, energy monitoring systems, and entertainment systems, among others. The proposed system offers several advantages, including a low-cost architecture, ease of setup, improved interoperability, high flexibility, and a lightweight frame that can be applied to other wireless-based smart systems and applications. Full article

Homegardens are traditional agroforestry systems that harbor genetic resources and ancestral knowledge, as well as contributing to food security and self-sufficiency in many rural communities. In this study, we analyze homegardens in a Mixtec community in coastal Oaxaca, Mexico, to document their arrangement and components, the useful flora and fauna they contain, and the social, cultural and economic aspects associated with their management. We used snowball sampling to perform semistructured interviews with 36 women in charge of homegardens, which represented 10% of the total homes in the community. During guided tours, we diagrammed the homegardens and collected and identified plant specimens to compile a full floristic listing. Plant specimens were deposited in the CHAP herbarium. We also calculated the Jacknife alpha diversity index and Sorensen’s beta diversity index to quantify the diversity of the garden flora. We summarized the interview data using descriptive statistics and performed a multiple regression analysis to evaluate the effects of the size of the homegarden and the homegarden owner’s age, years of school attendance, and language use on the number of useful plant species in the garden. Additionally, we conducted a multiple correspondence analysis on the homegardens, the sociodemographic variables, and the plant species contained. The components of the homegardens were the main dwelling, patio, kitchen, bathroom, chicken coop, and pigpen. We documented 15 animal species from 15 genera and 13 families and 236 plant species from 197 genera and 84 families. The most represented plant families were Araceae, Fabaceae and Apocynaceae. The main plant uses were ornamental, edible, and medicinal. The multiple correspondence analysis and multiple regression both showed sociodemographic variables to make a very low contribution to homegarden species richness (evidenced by low percentage variance explained and no statistically significant effects, respectively). The first-order Jacknife diversity index estimated a total of 309 plant species present in the homegardens, indicating high agrobiodiversity. The Sorensen index value ranged from 0.400 to 0.513. Similarity among the gardens was mostly due to high similarity among edible plants. There was community-level resilience in family food self-sufficiency, as 80.56% of the interviewees use harvest from their homegardens to cover their families’ food needs. Women play a central role in the establishment and management of the gardens. Overall, our findings demonstrate that homegardens in this community are sustainable; have high agrobiodiversity; provide food, medicine, and well-being to residents; contribute to food self-sufficiency; and conserve agrobiodiversity as well as traditional culture and knowledge. Full article

The rapid proliferation of Internet of Things (IoT) devices has increased the attack surface for cyber threats. Traditional intrusion detection systems often struggle to keep pace with novel or evolving threats. This study proposes an end-to-end generative AI-based intrusion detection and response pipeline designed for automated threat mitigation in smart home IoT environments. It leverages a Variational Autoencoder (VAE) trained on benign traffic to flag anomalies, a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) model to classify anomalies into five attack categories (C&C, DDoS, Okiru, PortScan, and benign), and Grok3—a large language model—to generate tailored countermeasure recommendations. Using the Aposemat IoT-23 dataset, the VAE model achieves a recall of 0.999 and a precision of 0.961 for anomaly detection. The BERT model achieves an overall accuracy of 99.90% with per-class F1 scores exceeding 0.99. End-to-end prototype simulation involving 10,000 network traffic samples demonstrate a 98% accuracy in identifying cyber attacks and generating countermeasures to mitigate them. The pipeline integrates generative models for improved detection and automated security policy formulation in IoT settings, enhancing detection and enabling quicker and actionable security responses to mitigate cyber threats targeting smart home environments. Full article

This article addresses the lack of low-cost, secure image-transmission solutions for IoT systems in remote environments. The design and implementation of a complete LoRa-based transmission system using ESP32 microcontrollers and Ebyte E220 modules, featuring AES-CBC encryption, HMAC integrity protection, and a custom retransmission protocol, are presented. The system achieves 100% packet delivery ratio (PDR) for 20 kB images over distances exceeding 2 km under line-of-sight conditions, with functional transmission up to 4.1 km. Image transmission time ranges from 35 s (0.1 m) to 110 s (600 m), while energy consumption increases from 4.95 mWh to 15.18 mWh. Critically, encryption imposes less than 1% overhead on total energy consumption. Unlike prior work focusing on isolated components, this article provides a complete, deployable architecture combining (i) low-cost hardware (<USD 50 total), (ii) long-range LoRa communication, (iii) custom reliability mechanisms for fragmenting 20 kB images into 100 packets, and (iv) end-to-end cryptographic protection, all evaluated experimentally across multi-kilometer distances. These findings demonstrate that secure long-range image transmission using commodity hardware is feasible and scalable for smart home and industrial monitoring applications. Full article

Precision agriculture increasingly recognizes the transformative potential of unmanned aerial systems (UASs) for crop monitoring and field assessment, yet research consistently highlights significant usability barriers as the main constraints to widespread adoption. Complex mission planning processes, including detailed flight plan creation and way point management, pose substantial technical challenges that mainly affect non-expert operators. Farmers and their teams generally prefer user-friendly, straightforward tools, as evidenced by the rapid adoption of GPS guidance systems, which underscores the need for simpler mission planning in UAS operations. To enhance accessibility and safety in UAS control, especially for non-expert operators in agriculture and related fields, we propose a Secure UAS Control Framework (SAUCF): a comprehensive system for natural-language-driven UAS mission management with integrated dual-factor biometric authentication. The framework converts spoken user instructions into executable flight plans by leveraging a language-model-powered mission planner that interprets transcribed voice commands and generates context-aware operational directives, including takeoff, location monitoring, return-to-home, and landing operations. Mission orchestration is performed through a large language model (LLM) agent, coupled with a human-in-the-loop supervision mechanism that enables operators to review, adjust, or confirm mission plans before deployment. Additionally, SAUCF offers a manual override feature, allowing users to assume direct control or interrupt missions at any stage, ensuring safety and adaptability in dynamic environments. Proof-of-concept demonstrations on a UAS plat-form with on-board computing validated reliable speech-to-text transcription, biometric verification via voice matching and face authentication, and effective Sim2Real transfer of natural-language-driven mission plans from simulation environments to physical UAS operations. Initial evaluations showed that SAUCF reduced mission planning time, minimized command errors, and simplified complex multi-objective workflows compared to traditional waypoint-based tools, though comprehensive field validation remains necessary to confirm these preliminary findings. The integration of natural-language-based interaction, real-time identity verification, human-in-the-loop LLM orchestration, and manual override capabilities allows SAUCF to significantly lower the technical barrier to UAS operation while ensuring mission security, operational reliability, and operator agency in real-world conditions. These findings lay the groundwork for systematic field trials and suggest that prioritizing ease of operation in mission planning can drive broader deployment of UAS technologies. Full article

Population aging is increasing dementia care demand. We present an audio-driven monitoring pipeline that operates either on mobile phones, microcontroller nodes, or smart television sets. The system combines audio signal processing with AI tools for structured interpretation. Preprocessing includes voice activity detection, speaker diarization, automatic speech recognition for dialogs, and speech-emotion recognition. An audio classifier detects home-care–relevant events (cough, cane taps, thuds, knocks, and speech). A large language model integrates transcripts, acoustic features, and a consented household knowledge base to produce a daily caregiver report covering orientation/disorientation (person, place, and time), delusion themes, agitation events, health proxies, and safety flags (e.g., exit seeking and falling). The pipeline targets real-time monitoring in homes and facilities, and it is an adjunct to caregiving, not a diagnostic device. Evaluation focuses on human-in-the-loop review, various audio/speech modalities, and the ability of AI to integrate information and reason. Intended users are low-income households in remote settings where in-person caregiving cannot be secured, enabling remote monitoring support for older adults with dementia. Full article

The dual Hukou system, originating in China’s planned economy period, structured Chinese society into separate urban and rural segments, thereby generating distinct sets of rights and benefits for agricultural and non-agricultural residents regarding land, social security, education, and healthcare. Urban home purchase is a pivotal indicator of social integration for rural–urban migrants in destination cities. While the literature has extensively examined migrants’ residential conditions in China, the institutional impact of the agricultural hukou system—a core constraint—on their urban homeownership, along with its underlying mechanisms and heterogeneity, remains underexplored. To address this gap, this study adopts a twofold approach: theoretically, it employs the separating equilibrium model in housing markets with incomplete information to verify that agricultural hukou acts as an institutional barrier to migrants’ local home purchases; empirically, it uses data from the China Migrants Dynamic Survey (CMDS) and applies the Fairlie decomposition method to quantify the constraint effect. The empirical results suggest that agricultural hukou exerts a 29.72% suppressive effect on migrants’ urban home purchase behavior. This effect operates indirectly by weakening migrants’ long-term settlement intention, which serves as a mediating variable. Moreover, the hindrance of agricultural hukou varies heterogeneously across groups, differing in education level, generational cohort, and regional distribution. To advance the fair and sustainable development of the real estate market, we advocate accelerating hukou reform by decoupling public services from residence status, fostering inclusive urbanization, and ensuring equitable development of housing markets. Full article

Snoring, a common sleep-disordered breathing phenomenon, impairs sleep quality for both the sufferer and any bed partner. While mild snoring primarily disrupts sleep continuity, severe cases often indicate obstructive sleep apnea (OSA), a disorder affecting 9–17% of the global population, linked to significant comorbidities and socioeconomic burden (see Introduction for supporting data). Here, we propose a low-cost, real-time snoring detection and intervention system that integrates a multiple-spectrum deep learning framework with an Internet of Things (IoT)-enabled smart pillow. The modified Parallel Convolutional Spatiotemporal Network (PCSN) combines three parallel convolutional neural network (CNN) branches processing Constant-Q Transform (CQT), Synchrosqueezing Wavelet Transform (SWT), and Hilbert–Huang Transform (HHT) features with a Long Short-Term Memory (LSTM) network to capture spatial and temporal characteristics of sounds associated with snoring. The smart pillow prototype incorporates two Micro-Electro-Mechanical System (MEMS) microphones, an ESP8266 off-shelf board, a speaker, and two vibration motors for real-time audio acquisition, cloud-based processing via Arduino cloud, and closed-loop haptic/audio feedback that encourages positional changes without fully awakening the snorers. Experiments demonstrated that the modified PCSN model achieves 98.33% accuracy, 99.29% sensitivity, 98.34% specificity, 98.3% recall, and 98.32% F1-score, outperforming existing systems. Hardware costs are under USD 8 and a smartphone app provides authorized users with real-time visualization and secure data access. This solution offers a cost-effective and accurate approach for home-based OSA screening and intervention. Full article

By 2050, most adults aged 65 and older in the United States will want to age independently at home, a goal that will strain healthcare resources. Adults aged 50 and older (N = 112) were recruited for study participation between 2018 and 2022. They completed surveys and participated in discussion sessions to explore their needs and opinions regarding smart home sensors. Survey results indicated that older adults’ comfort with smart home sensors increased with their perceived need for monitoring when home alone (OR = 1.46; p = 0.012) or sick/recovering from an illness (OR = 2.21; p < 0.001). When sick compared to when healthy, individuals were 2.65 times more likely to prefer installing multiple sensors in the living room, 1.75 times more likely in the kitchen, 3.66 times more likely in the bedroom, and 3.41 times more likely in the bathroom (p < 0.05). Regarding data sharing, participants were most willing to share information with healthcare providers and family members on a regular basis (80 and 81%, respectively) and 71% on a regular basis or when sick/recovering. Comfort with data sharing with professional caregivers (OR = 1.67; p = 0.0017) and monitoring companies (OR = 1.34; p = 0.030) significantly increased when sick/recovering. Discussion sessions highlighted overwhelming concerns about personal security/privacy, loss of independence, and ethical issues in data collection. Participants emphasized the need for new systems to be flexible, cost-effective, user-friendly, and respectful of user autonomy, accommodating diverse life stages, comfort levels, home environments, income levels, and support structures. Insights are now informing sensor data collection in our model home. Study findings underscore the importance of involving potential users in technology development to create effective and acceptable solutions for aging in place. Full article

This paper presents a comprehensive methodology to forecast the daily energy demand associated with recharging private electric vehicles in urban areas. The approach is based on plausible scenarios regarding the penetration of battery-powered vehicles and the availability of charging infrastructure. Accurate space and time forecasting of charging activities and power requirements is a critical issue in supporting the transition from conventional to battery-powered vehicles for urban mobility. This technological shift represents a key milestone toward achieving the zero-emissions target set by the European Green Deal for 2050. The methodology leverages Floating Car Data (FCD) samples. The widespread use of On-Board Units (OBUs) in private vehicles for insurance purposes ensures the methodology’s applicability across diverse geographical contexts. In addition to FCD samples, the estimation of charging demand for private electric vehicles is informed by a large-scale, detailed survey conducted by ENEA in Italy in 2023. Funded by the Ministry of Environment and Energy Security as part of the National Research on the Electric System, the survey explored individual charging behaviors during daily urban trips and was designed to calibrate a discrete choice model. To date, the methodology has been applied to the Metropolitan Area of Rome, demonstrating robustness and reliability in its results on two different scenarios of analysis. Each demand/supply scenario has been evaluated in terms of the hourly distribution of peak charging power demand, at the level of individual urban zones or across broader areas. Results highlight the role of the different components of power demand (at home or at other destinations) in both scenarios. Charging at intermediate destinations exhibits a dual peak pattern—one in the early morning hours and another in the afternoon—whereas home-based charging shows a pronounced peak during evening return hours and a secondary peak in the early afternoon, corresponding to a decline in charging activity at other destinations. Power distributions, as expected, sensibly differ from one scenario to the other, conditional to different assumptions of private and public recharge availability and characteristics. Full article

This study presents the design and validation of an open-source framework for biomedical signal acquisition and interoperable data exchange based on the Health Level Seven—Fast Healthcare Interoperability Resources (HL7 FHIR) standard. The proposed system enables secure, wireless transmission of physiological data from distributed sensing nodes toward a locally hosted monitoring platform. The hardware architecture integrates ESP32-WROOM-32 microcontrollers for multi-parameter acquisition, the MQTT protocol for low-latency communication, and a Home Assistant (Nabu Casa, San Diego, CA, USA)–InfluxDB (InfluxData, San Francisco, CA, USA)–Grafana (Grafana Labs, New York, NY, USA) stack for real-time visualization. The novelty of this work lies in the full-stack implementation of HL7 FHIR Observations within a reproducible, open-source environment, ensuring semantic interoperability without reliance on proprietary middleware or cloud services. A case study involving multi-sensor acquisition of electrocardiographic (ECG), photoplethysmographic (PPG), temperature, and oxygen saturation signals was conducted to evaluate system performance. Validation results confirmed consistent end-to-end data flow, sub-second latency, zero packet loss, and accurate semantic preservation across all processing stages. These findings demonstrate the feasibility of implementing standardized, open, and scalable biomedical Internet of Medical Things (IoMT) systems using non-proprietary components. The proposed framework provides a reproducible foundation for future telemedicine and continuous patient-monitoring applications, aligning with FAIR data principles and the ongoing digital transformation of healthcare. Full article

The necessity for intelligent monitoring has grown more urgent as the number of crimes involving firearms and knives in homes and public areas has increased. Traditional CCTV systems require human operators, whose attentiveness and the impracticability of monitoring multiple video feeds simultaneously limit their effectiveness. Artificial intelligence (AI)-based vision systems can automatically detect firearms and enhance public safety, thereby overcoming this constraint. In accordance with the Preferred Reporting Items for Systematic Reviews (PRISMA) criteria, a systematic evaluation of AI-based weapon detection for security monitoring is conducted. The paper summarizes research works on AI, machine learning, and deep learning techniques for identifying weapons in surveillance footage from 2016 to 2025, encompassing 101 research papers. The reported precision ranged from 78% to 99.5%, recall ranged from 83% to 97%, and mean average precision (mAP) ranged from approximately 70% to 99%. While AI-based monitoring significantly enhances detection accuracy, issues with inconsistent evaluation criteria, limited real-world validation, and dataset variability persist. The research study emphasizes the need for uniform benchmarking, robust privacy protections, and standardized datasets to ensure the ethical and reliable implementation of AI-driven weapon-detection systems. Full article

The proliferation of the Social Internet of Things (SIoT) necessitates robust and scalable trust management systems to ensure secure and reliable interactions among heterogeneous devices. However, existing trust management models often lack scalability for large SIoT environments. To address this, a lightweight trust evaluation model for SIoT, referred to as Micro-Moment (MMTE), is presented here. MMTE evaluates trust based on concise, context specific, repetitive, and high-frequency interactions, termed micro-moments among SIoT devices. The MMTE model is evaluated using the Lysis dataset, which is extracted from a real SIoT environment, and demonstrates superior resource efficiency compared to existing SIoT trust models with significantly lower CPU time, memory, and disk usage. MMTE’s linear complexity and simple design make it more resource efficient and scalable than other lightweight trust models, especially when processing large-scale data in heterogeneous SIoT networks. Moreover, MMTE accurately distinguishes 99.35% of malicious nodes in a simulated smart home environment. Furthermore, a numerical comparison clearly demonstrates that MMTE outperforms existing and recently published trust models in terms of classifying malicious and benign nodes. To enhance scalability, the concept of trust spheres is introduced, and devices with similar trust scores are grouped to streamline processing and storage demands. Sphere Anchors manage the trust spheres and efficiently distribute computational tasks and optimize storage through an adaptive storage strategy. The trust spheres also efficiently manage increasing network sizes, maintaining linear processing times as the traffic load increases, and also outperform existing models in terms of average propagation times. MMTE and trust spheres together provide a robust, scalable, and lightweight solution for trust management in SIoT networks. Full article

Cloud-centric IoT frameworks remain dominant; however, they introduce major challenges related to data privacy, latency, and system resilience. Existing open-source solutions often lack standardized principles for scalable, local-first deployment and do not adequately integrate fault tolerance with hybrid automation logic. This study presents a practical and extensible local-first IoT architecture designed for full operational autonomy using open-source components. The proposed system features a modular, layered design that includes device, communication, data, management, service, security, and presentation layers. It integrates MQTT, Zigbee, REST, and WebSocket protocols to enable reliable publish–subscribe and request–response communication among heterogeneous devices. A hybrid automation model combines rule-based logic with lightweight data-driven routines for context-aware decision-making. The implementation uses Proxmox-based virtualization with Home Assistant as the core automation engine and operates entirely offline, ensuring privacy and continuity without cloud dependency. The architecture was deployed in a real-world office environment and evaluated under workload and fault-injection scenarios. Results demonstrate stable operation with MQTT throughput exceeding 360,000 messages without packet loss, automatic recovery from simulated failures within three minutes, and energy savings of approximately 28% compared to baseline manual control. Compared to established frameworks such as FIWARE and IoT-A, the proposed approach achieves enhanced modularity, local autonomy, and hybrid control capabilities, offering a reproducible model for privacy-sensitive smart environments. Full article