Internet of Things最新文献_第9页

Digital twin-driven semantic offloading for LEO-MEC-enabled remote IoT networks 支持leo - mec的远程物联网网络的数字双驱动语义卸载

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-21 DOI: 10.1016/j.iot.2025.101831

Ayalneh Bitew Wondmagegn, Dongwook Won, Donghyun Lee, Jaemin Kim, Juyoung Kim, Sungrae Cho

{"title":"Digital twin-driven semantic offloading for LEO-MEC-enabled remote IoT networks","authors":"Ayalneh Bitew Wondmagegn, Dongwook Won, Donghyun Lee, Jaemin Kim, Juyoung Kim, Sungrae Cho","doi":"10.1016/j.iot.2025.101831","DOIUrl":"10.1016/j.iot.2025.101831","url":null,"abstract":"<div><div>Efficient task offloading to low-Earth orbit (LEO) satellite-assisted mobile edge computing (MEC) servers is vital for addressing the computational delay and energy constraints of remote Internet of Things (IoT) devices. However, inherent challenges in LEO satellite networks–including dynamic topology, intermittent connectivity, and bandwidth scarcity undermine the performance of conventional offloading schemes. Moreover, raw data transmission inflates latency and energy consumption, hindering real-time and resource-aware decision-making. To overcome these limitations, this paper presents a digital twin (DT)-driven, semantic-aware offloading framework tailored for LEO-integrated MEC systems. The proposed framework aims to minimize end-to-end DT synchronization delay while adhering to strict energy, semantic fidelity, and computational constraints. Semantic encoders extract task-relevant information to compress payloads, and a binary offloading strategy ensures each task is either fully executed locally or offloaded entirely. The problem is formulated as a joint optimization of offloading decisions, edge server selection, semantic compression ratio, and transmission power allocation. A proximal policy optimization (PPO)-based deep reinforcement learning algorithm is developed to solve the problem, leveraging real-time DT feedback for adaptive, context-aware control under dynamic network conditions. Simulation results demonstrate that the proposed framework reduces DT synchronization delay by up to 45 % vs. local execution, by 26–35 % vs. non-semantic offloading, and by 5–10 % vs. DRL alternative at high load, with statistically significant gaps. Additionally, the system maintains energy and semantic fidelity requirements while scaling efficiently with data volume and device density. These findings offer a practical and scalable solution for enabling reliable, low-latency DT services in 6G satellite–ground integrated IoT networks.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101831"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FTL-TSLP: A federated transfer learning approach with a two-stage LSTM pipeline for fault-tolerant and privacy-preserving intrusion detection in IoMT networks FTL-TSLP：一种基于两阶段LSTM管道的联合迁移学习方法，用于IoMT网络中的容错和隐私保护入侵检测

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-23 DOI: 10.1016/j.iot.2025.101832

Abdelhammid Bouazza , Hichem Debbi , Hicham Lakhlef

{"title":"FTL-TSLP: A federated transfer learning approach with a two-stage LSTM pipeline for fault-tolerant and privacy-preserving intrusion detection in IoMT networks","authors":"Abdelhammid Bouazza , Hichem Debbi , Hicham Lakhlef","doi":"10.1016/j.iot.2025.101832","DOIUrl":"10.1016/j.iot.2025.101832","url":null,"abstract":"<div><div>The rapid proliferation of the Internet of Medical Things (IoMT) has transformed healthcare delivery by enabling continuous patient monitoring, intelligent clinical decision-making, and efficient remote care. However, these advancements have also introduced substantial cybersecurity risks that threaten patient privacy, safety, and the operational resilience of healthcare systems. These challenges are further compounded by stringent regulatory requirements and the inherent complexity of heterogeneous, non-independent, and identically distributed (non-IID) data. To address these challenges, we propose FTL-TSLP, a novel federated intrusion detection framework that integrates federated learning (FL) with targeted transfer learning (TL) through a two-stage LSTM-based pipeline. The framework is explicitly designed to operate effectively under both IID and non-IID data distributions while preserving data privacy. On the client side, temporal aggregation techniques efficiently compress sequential data, reducing computational costs without compromising detection accuracy. Additionally, the framework enhances fault tolerance by incorporating a Multi-Criteria Decision Analysis (MCDA) module combined with a Naïve Bayes classifier for real-time, probabilistic device-level classification. The proposed model demonstrates superior performance across the NF-UNSW-NB15-v2, WUSTL-EHMS-2020, and CICIoMT-2024 benchmark datasets. Even under extreme Dirichlet-based non-IID conditions (<span><math><mrow><mi>α</mi><mo>=</mo><mn>0.1</mn></mrow></math></span>), FTL-TSLP achieves 99.72 % accuracy and a 98.07 % F1-score on the CICIoMT-2024 dataset, confirming its robustness in heterogeneous IoMT traffic environments. These results highlight that FTL-TSLP offers a reliable, privacy-preserving, and computationally efficient solution for securing IoMT healthcare ecosystems.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101832"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Blockchain-enabled traceability for rice blending fraud: A practical framework to strengthen supply chain integrity in smart agricultural IoT 支持区块链的大米混合欺诈可追溯性：智能农业物联网中加强供应链完整性的实用框架

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-14 DOI: 10.1016/j.iot.2025.101824

Tao Peng, Langfeng Wang

引用次数: 0

PENSIL: Programmable network stack for low-power lossy IoT networks using lightweight-virtualization PENSIL：可编程网络堆栈，用于使用轻量级虚拟化的低功耗物联网网络

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-24 DOI: 10.1016/j.iot.2025.101829

Ahmad Mahmod , Julien Montavont , Thomas Noel

{"title":"PENSIL: Programmable network stack for low-power lossy IoT networks using lightweight-virtualization","authors":"Ahmad Mahmod , Julien Montavont , Thomas Noel","doi":"10.1016/j.iot.2025.101829","DOIUrl":"10.1016/j.iot.2025.101829","url":null,"abstract":"<div><div>Low-Power and Lossy Wireless Networks (LLWNs) form the foundation of the Internet of Things (IoT), connecting billions of constrained devices across diverse domains. Despite their critical role, the design of LLWN devices is strongly constrained by limited memory, processing power, and energy supply. These limitations have historically led to the adoption of monolithic network stacks, where protocol logic is tightly integrated and bound at compile time. As a result, even minor changes require a full firmware update, making protocol evolution costly and impractical. Because LLWN deployments face diverse and evolving conditions, a single static stack design or fixed configuration is insufficient. In this paper, we propose PENSIL, a network architecture featuring a programmable and modular network stack for LLWN that enables selective updates of protocol functions, combined with a central orchestrator that manages device stacks. PENSIL enables dynamic and semantic reconfiguration, from parameter tuning to network configuration swapping, allowing networks to adapt without downtime. A proof-of-concept implementation on real hardware demonstrates that our architecture enhances performance through fast, lightweight and secure updates while respecting the stringent memory, energy, and processing constraints of LLWN devices, ultimately bridging the gap between programmability and efficiency.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101829"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Collective intelligence-based service migration enabling zoom-in functionality within industry 5.0 基于集体智能的服务迁移，支持工业5.0中的放大功能

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-26 DOI: 10.1016/j.iot.2025.101830

Riccardo Venanzi , Lorenzo Colombi , Davide Tazzioli , Simon Dahdal , Mauro Tortonesi , Luca Foschini

{"title":"Collective intelligence-based service migration enabling zoom-in functionality within industry 5.0","authors":"Riccardo Venanzi , Lorenzo Colombi , Davide Tazzioli , Simon Dahdal , Mauro Tortonesi , Luca Foschini","doi":"10.1016/j.iot.2025.101830","DOIUrl":"10.1016/j.iot.2025.101830","url":null,"abstract":"<div><div>The rapid evolution of Industry 5.0 emphasizes the integration of human expertise with machine intelligence to create resilient, adaptive, and human-centric industrial systems. This paper introduces a novel Collective Intelligence (CI)-based service migration framework designed for Industry 5.0 environments, enabling dynamic orchestration of stateful services across heterogeneous edge-to-cloud infrastructures. At its core, the framework leverages Kubernetes (K8s) enhanced with AI-driven decision-making and human-in-the-loop collaboration to address the limitations of traditional orchestration in industrial settings. A key innovation of this work is the Zoom-In functionality, which empowers human operators to escalate anomaly detection and analysis by deploying advanced machine learning models on demand, seamlessly migrating services to resource-rich nodes when deeper investigation is warranted. The proposed framework integrates Large Language Models (LLMs) to translate operator intent into actionable policies, ensuring context-aware and explainable decision-making. Experimental validation in real industrial scenarios demonstrates high anomaly detection accuracy (F1-scores up to 1.0), reliable operator intent translation (over 70 % correct JSON generations with lightweight LLMs), and efficient multi-criteria scheduling with millisecond-level decision times. Moreover, the proposed migration mechanism reduces downtime by more than 50 % compared to vanilla Kubernetes, ensuring service continuity in mission-critical tasks. This work advances the vision of collaborative intelligence in IoT systems, bridging the gap between human judgment and automated orchestration for Industry 5.0 applications.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101830"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AtomicVAD: A tiny voice activity detection model for efficient inference in intelligent IoT systems AtomicVAD：用于智能物联网系统中高效推理的微型语音活动检测模型

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-20 DOI: 10.1016/j.iot.2025.101822

Angelo J. Soto-Vergel , Prashant Sankaran , Juan C. Velez , Rene Amaya-Mier , Diana Ramirez-Rios

{"title":"AtomicVAD: A tiny voice activity detection model for efficient inference in intelligent IoT systems","authors":"Angelo J. Soto-Vergel , Prashant Sankaran , Juan C. Velez , Rene Amaya-Mier , Diana Ramirez-Rios","doi":"10.1016/j.iot.2025.101822","DOIUrl":"10.1016/j.iot.2025.101822","url":null,"abstract":"<div><div>This paper introduces AtomicVAD, an ultra-lightweight, end-to-end voice activity detection (VAD) model designed for inference on resource-constrained microcontrollers at the extreme edge. Existing VAD models often rely on large architectures with thousands of trainable parameters, making them impractical for deployment on low-power microcontrollers commonly used in internet of things systems. Even with compression methods such as quantization or pruning, these models typically fail to achieve low-latency performance under strict power and memory limits. AtomicVAD overcomes these limitations through the introduction of the General Growing Cosine Unit, a trainable oscillatory activation function that embeds feature learning within periodic modulations. This design enables remarkable efficiency with approximately 0.3k trainable parameters, representing a 99.7 % reduction compared to commonly used baselines like MarbleNet, while maintaining competitive accuracy. Evaluated on the challenging AVA-Speech benchmark, AtomicVAD achieves an AUROC of 0.903 and an F<sub>2</sub>-score of 0.891, outperforming larger state-of-the-art systems and demonstrating robustness to background noise and music. Optimized for extreme efficiency, AtomicVAD enables ultra-low latency inference —as low as 26 ms on a 240 MHz Cortex-M7 and 1.22 s on a 64 MHz Cortex-M4F— facilitated by INT8 quantization. Its memory footprint remains below 75 kB Flash and 65 kB SRAM. A real-world LoRaWAN field trial further validated its practicality, showing that on-device speech gating eliminates unnecessary, bandwidth-intensive audio uploads, reducing over-the-air delays from minutes to milliseconds. Key use cases include remote monitoring, smart-home control, disaster-response sensor networks, and other long-range, low-power systems requiring efficient, always-on audio processing.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101822"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A privacy-aware and sustainable joint optimization for resource-constrained internet of things using deep reinforcement learning 基于深度强化学习的资源受限物联网隐私感知可持续联合优化

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-12-01 DOI: 10.1016/j.iot.2025.101837

Mehdi Hosseinzadeh , Parisa Khoshvaght , Amir Masoud Rahmani , Farhad Soleimanian Gharehchopogh , Shakiba Rajabi , Aso Darwesh , Omed Hassan Ahmed , Thantrira Porntaveetus , Sang-Woong Lee

{"title":"A privacy-aware and sustainable joint optimization for resource-constrained internet of things using deep reinforcement learning","authors":"Mehdi Hosseinzadeh , Parisa Khoshvaght , Amir Masoud Rahmani , Farhad Soleimanian Gharehchopogh , Shakiba Rajabi , Aso Darwesh , Omed Hassan Ahmed , Thantrira Porntaveetus , Sang-Woong Lee","doi":"10.1016/j.iot.2025.101837","DOIUrl":"10.1016/j.iot.2025.101837","url":null,"abstract":"<div><div>The rise of battery-powered Internet of Thing (IoT) fleets in buildings and campuses requires policies that manage sensing, communication, and edge–cloud offloading while considering energy, carbon, privacy, and cost limits. In this paper, we frame this challenge as a Markov Decision Process (MDP) and design a controller using Deep Reinforcement Learning (DRL). We present a Rainbow-based IoT controller that retains distributional value learning, dueling networks, NoisyNets, <em>n</em>-step returns, Prioritized Experience Replay (PER), and double selection, and contributes four novelties: dual-budget Lagrangian control with warm-up, connectivity-robust distributional targets reweighted by outage/queue risk, federated sketch-guided replay for underrepresented regimes, and realistic ISAC-aware macro-actions with integrated DP/CO₂ accounting and budget-aware training/logging. Simulations show that the proposed algorithm achieves ≈88 % higher anomaly detection, ≈39 % higher packet success, ≈52 % less energy consumption, and ≈74 % lower cloud cost than the best baseline, demonstrating superior utility, reliability, and sustainability in IoT workloads.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101837"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IoT-enabled CCTV monitoring and deep learning for automated water body segmentation in agricultural reservoirs 基于物联网的闭路电视监控和深度学习，用于农业水库的自动水体分割

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-12 DOI: 10.1016/j.iot.2025.101823

Soon Ho Kwon , Youcan Feng , Seungyub Lee

{"title":"IoT-enabled CCTV monitoring and deep learning for automated water body segmentation in agricultural reservoirs","authors":"Soon Ho Kwon , Youcan Feng , Seungyub Lee","doi":"10.1016/j.iot.2025.101823","DOIUrl":"10.1016/j.iot.2025.101823","url":null,"abstract":"<div><div>This study presents an IoT-enabled deep learning framework for automated water body segmentation in agricultural reservoirs using CCTV imagery. The framework integrates IoT-based long-term monitoring imagery with public benchmark water datasets to improve segmentation robustness under diverse visual conditions encountered in real-world agricultural settings. Three training strategies were defined—(I) benchmark-only, (II) CCTV-only, and (III) integrated—to systematically evaluate generalization across heterogeneous data sources. We also evaluate three U-Net–based architectures: Model 1 (baseline with binary cross-entropy loss), Model 2 (the same architecture trained with a differentiable Jaccard loss), and Model 3 (a parameter-reduced architecture with weighted cross-entropy, designed for edge inference on IoT hardware). Each configuration was trained and validated over ten independent runs to ensure statistical reliability. Model 2 consistently achieved the highest and most stable performance across all training strategies, demonstrating that loss-function optimization, rather than architectural expansion alone, is the primary driver of performance improvement. The integrated training strategy (Strategy III), which combines benchmark and CCTV imagery, yielded the strongest generalization, improving mean IoU by roughly 15–20% and reducing variability compared to single-source training. Performance differences between test sites were attributable mainly to environmental variability—fog, reflections, shadows, surface ripples, and vegetation occlusions—rather than model instability. Validated on two independent reservoirs that were fully held out from training, the framework generalized to new sites without requiring additional site-specific annotation. Model 3′s parameter-efficient design (approximately 60% fewer trainable parameters) supports near-edge inference on embedded IoT devices, enabling continuous, unattended, IoT-based monitoring of agricultural reservoirs for smart irrigation management.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101823"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An efficient real-time apple disease classification approach using a novel lightweight neural network on an Arduino edge device 在Arduino边缘设备上使用新型轻量级神经网络的高效实时苹果病害分类方法

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-11-22 DOI: 10.1016/j.iot.2025.101833

Milan Grujev , Dragan Stojanovic , Aleksandar Milosavljevic , Milos Ilic , Veljko Prodanovic

{"title":"An efficient real-time apple disease classification approach using a novel lightweight neural network on an Arduino edge device","authors":"Milan Grujev , Dragan Stojanovic , Aleksandar Milosavljevic , Milos Ilic , Veljko Prodanovic","doi":"10.1016/j.iot.2025.101833","DOIUrl":"10.1016/j.iot.2025.101833","url":null,"abstract":"<div><div>Advances in miniaturization, battery technology, and software enable edge devices to capture high-resolution images and videos, perform real-time processing, and support faster decision-making in precision agriculture. Automated apple disease detection is crucial for maintaining crop health and preventing large-scale loss. Conventional methods rely on large neural networks that require high-performance computing resources, making them impractical for real-time on-site deployment. These computational demands limit their usability in agricultural settings, where power and connectivity constraints are common. This research aims to develop a novel, optimized machine learning model for apple leaf disease classification, designed specifically for low-power edge devices. Based on core principles of the MobileNet architecture, the model employs depthwise convolution blocks, a layer reduction process, and memory-efficient optimization techniques to achieve high accuracy while significantly reducing computational overhead. Trained on a combination of readily available datasets, the model achieved a classification accuracy of 94.80 %, with a model size of just 36.2 KB and runtime memory usage of 162.2 KB, representing a reduction of over 90 % compared to the standard MobileNetV2. The model performed inference in 463.8 milliseconds while consuming only 14 milliwatts of power. These results demonstrate the potential of lightweight neural networks for real-time disease detection on resource-constrained devices such as Arduino microcontrollers. By enabling localized, on-device inference without cloud dependency, Edge AI offers a scalable and cost-effective solution for precision agriculture, improving crop monitoring and sustainability.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101833"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145615719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing precision irrigation with TinyML: Advanced NDVI anomaly detection and model optimization 用TinyML增强精准灌溉：先进的NDVI异常检测和模型优化

IF 7.6 3区计算机科学

Internet of Things Pub Date : 2026-01-01 Epub Date: 2025-12-03 DOI: 10.1016/j.iot.2025.101839

Carlos Hernandez-Hidalgo, Aurora González-Vidal, Antonio F. Skarmeta

{"title":"Enhancing precision irrigation with TinyML: Advanced NDVI anomaly detection and model optimization","authors":"Carlos Hernandez-Hidalgo, Aurora González-Vidal, Antonio F. Skarmeta","doi":"10.1016/j.iot.2025.101839","DOIUrl":"10.1016/j.iot.2025.101839","url":null,"abstract":"<div><div>Agriculture accounts for over 70 % of global freshwater use, yet water availability is increasingly threatened by climate change, overuse, and poor management. Smallholder farmers are particularly vulnerable, often lacking access to advanced technologies. Sustainable agriculture demands innovative, energy-efficient solutions for smarter water management. This work proposes a low-power precision irrigation approach using Tiny Machine Learning (TinyML) to operate without cloud connectivity in resource-constrained environments and investigates how anomalies in the Normalized Difference Vegetation Index (NDVI) combined with environmental data such as temperature and humidity, can drive adaptive, data-driven irrigation. Different percentile thresholds (e.g., 25th–75th) were evaluated to optimize detection. Models were trained in Keras and quantized from 32-bit to 8-bit using TensorFlow Lite for deployment on microcontrollers, enabling real-time inference without internet access. Three models were compared: Linear Regression (CVRMSE = 30.16 %), Random Forest Regression (RMSE = 0.062, CVRMSE = 27.42 %), and a Neural Network (RMSE = 0.0589, CVRMSE = 36.88 %) designed for TinyML deployment. The Percentile-based NDVI Anomaly Index (PNAI) improved predictive performance by up to 56.84 % in CVRMSE over standard methods, with the 25th–75th percentile range yielding the most accurate results. After quantization, the TinyML neural network achieved an RMSE of 0.0421 and a CVRMSE of 33.41 %, with only a 1.2 % accuracy drop and a model size of 6280 bytes, confirming its feasibility for on-device execution. These results demonstrate that TinyML-based NDVI anomaly detection is a viable, low-cost, and scalable approach for precision irrigation, with future work focusing on multi-crop validation and real-world field deployment.</div></div>","PeriodicalId":29968,"journal":{"name":"Internet of Things","volume":"35 ","pages":"Article 101839"},"PeriodicalIF":7.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145681813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0