A review of deep learning techniques used in agriculture

Ecological Informatics, Год журнала: 2023, Номер 77, С. 102217 - 102217

Опубликована: Июль 18, 2023

Язык: Английский

---

IoT: Communication protocols and security threats

Internet of Things and Cyber-Physical Systems, Год журнала: 2023, Номер 3, С. 1 - 13

Опубликована: Янв. 1, 2023

In this study, we review the fundamentals of IoT architecture and thoroughly present communication protocols that have been invented especially for technology. Moreover, analyze security threats, general implementation problems, presenting several sectors can benefit most from development. Discussion over findings reveals open issues challenges specifies next steps required to expand support systems in a secure framework.

Язык: Английский

---

An optimized CNN-based intrusion detection system for reducing risks in smart farming

Internet of Things, Год журнала: 2023, Номер 22, С. 100709 - 100709

Опубликована: Фев. 4, 2023

Язык: Английский

---

Next-Generation Wheat Disease Monitoring: Leveraging Federated Convolutional Neural Networks for Severity Estimation

Опубликована: Май 26, 2023

In this study, we investigate using federated learning for the CNN model-based prediction of wheat disease severity levels. We employed safe aggregation approaches to training model on dispersed data while maintaining privacy and security data. The dataset consisted 8643 photos plants with 10 levels illness. assessed efficacy a variety assessment measures, such as accuracy, precision, recall, F1 score, AUC ROC, validation set. With an accuracy 0.92, precision 0.87, recall score 0.89, ROC 0.95, findings demonstrated that performed well across all criteria. effectiveness centralized trained same were also compared. To could perform best hyperparameters, difference less than 0.05. study indicates promise reliable method estimating model. ability keep device lowers danger breaches ensures user privacy. This is possible local each participating node methods. further highlight significance distribution thorough hyperparameter tweaking optimum performance. results stimulate more studies in field aid creating strategies machine several domains.

Язык: Английский

---

Internet of Things and Wireless Sensor Networks for Smart Agriculture Applications: A Survey

IEEE Access, Год журнала: 2023, Номер 11, С. 145813 - 145852

Опубликована: Янв. 1, 2023

The increasing food scarcity necessitates sustainable agriculture achieved through automation to meet the growing demand. Integrating Internet of Things (IoT) and Wireless Sensor Networks (WSNs) is crucial in enhancing production across various agricultural domains, encompassing irrigation, soil moisture monitoring, fertilizer optimization control, early-stage pest crop disease management, energy conservation. application protocols such as ZigBee, WiFi, SigFox, LoRaWAN are commonly employed collect real-time data for monitoring purposes. Embracing advanced technology imperative ensure efficient annual production. Therefore, this study emphasizes a comprehensive, future-oriented approach, delving into IoT-WSNs, wireless network protocols, their applications since 2019. It thoroughly discusses overview IoT WSNs, architectures summarization protocols. Furthermore, addresses recent issues challenges related IoT-WSNs proposes mitigation strategies. provides clear recommendations future, emphasizing integration aiming contribute future development smart systems.

Язык: Английский

---

Revolutionizing Cyber Threat Detection With Large Language Models: A Privacy-Preserving BERT-Based Lightweight Model for IoT/IIoT Devices

IEEE Access, Год журнала: 2024, Номер 12, С. 23733 - 23750

Опубликована: Янв. 1, 2024

The field of Natural Language Processing (NLP) is currently undergoing a revolutionary transformation driven by the power pre-trained Large Models (LLMs) based on groundbreaking Transformer architectures. As frequency and diversity cybersecurity attacks continue to rise, importance incident detection has significantly increased. IoT devices are expanding rapidly, resulting in growing need for efficient techniques autonomously identify network-based networks with both high precision minimal computational requirements. This paper presents SecurityBERT, novel architecture that leverages Bidirectional Encoder Representations from Transformers (BERT) model cyber threat networks. During training we incorporated privacy-preserving encoding technique called Privacy-Preserving Fixed-Length Encoding (PPFLE). We effectively represented network traffic data structured format combining PPFLE Byte-level Byte-Pair (BBPE) Tokenizer. Our research demonstrates SecurityBERT outperforms traditional Machine Learning (ML) Deep (DL) methods, such as Convolutional Neural Networks (CNNs) or Recurrent (RNNs), detection. Employing Edge-IIoTset dataset, our experimental analysis shows achieved an impressive 98.2% overall accuracy identifying fourteen distinct attack types, surpassing previous records set hybrid solutions GAN-Transformer-based architectures CNN-LSTM models. With inference time less than 0.15 seconds average CPU compact size just 16.7MB, ideally suited real-life suitable choice deployment resource-constrained devices.

Язык: Английский

---

2DF-IDS: Decentralized and differentially private federated learning-based intrusion detection system for industrial IoT

Computers & Security, Год журнала: 2023, Номер 127, С. 103097 - 103097

Опубликована: Янв. 10, 2023

Язык: Английский

---

Empowering Precision Agriculture: Detecting Apple Leaf Diseases and Severity Levels with Federated Learning CNN

Опубликована: Июнь 23, 2023

The production of apples contributes significantly to the world's food security, but it also confronts significant obstacles because illnesses that harm apple leaves. Early diagnosis and categorization are essential effectively managing controlling these diseases supporting sustainable farming. Convolutional neural networks (CNNs) have powerful image classification capabilities. This research paper introduces a novel method classify leaf into four severity levels using federated learning CNNs, aiming harness privacy-preserving nature decentralized training benefits learning. To replicate environment, we created large dataset 8,973 labelled photos representing range disease categories levels. We then disseminated data among clients. Our process allowed for efficient use various datasets privacy protection our CNN model. trial results show efficacy suggested strategy, with F1 scores ranging from 93.26% 95.94% accuracy values all customers between 96% 98%. These performance measures model can adequately manage availability issues while classifying diseases. adds knowledge on categorizing plant provides insightful information next precision agriculture research. highlights potential this approach applications in agricultural domain, paving way more effective solutions by demonstrating viability CNNs classification.

Язык: Английский

---

IoT Malware Analysis Using Federated Learning: A Comprehensive Survey

IEEE Access, Год журнала: 2023, Номер 11, С. 5004 - 5018

Опубликована: Янв. 1, 2023

The Internet of Things (IoT) has paved the way to a highly connected society where all things are interconnected and exchanging information become more accessible through internet. With use IoT devices, threat malware increased rapidly. number existing new variants made protecting devices networks challenging. can hide in systems disables its activity when there attempts discover detect them. technological advances, various emerging techniques address this problem. However, they still encounter issues concerning privacy security user's data suffer from single point failure. To issue, recent research developments conducted Federated Learning (FL). FL is decentralized technique that trains on-device exchanges parameters without sharing data. implemented secure data, provide safe accurate models, prevent failure centralized models. This paper provides an overview different approaches integrate with IoT. Finally, we discuss applications FL, challenges, future directions.

Язык: Английский

---

Security of Internet of Things (IoT) using federated learning and deep learning — Recent advancements, issues and prospects

ICT Express, Год журнала: 2023, Номер 9(5), С. 941 - 960

Опубликована: Март 21, 2023

There is a great demand for an efficient security framework which can secure IoT systems from potential adversarial attacks. However, it challenging to design suitable model considering the dynamic and distributed nature of IoT. This motivates researchers focus more on investigating role machine learning (ML) in designing models. A brief analysis different ML algorithms discussed along with advantages limitations algorithms. Existing studies state that suffer problem high computational overhead risk privacy leakage. In this context, review focuses implementation federated (FL) deep (DL) security. Unlike conventional techniques, FL models maintain data while sharing information other systems. The study suggests overcome drawbacks techniques terms maintaining discusses models, overview, comparisons, summarization DL-based

Язык: Английский

---