Federated Learning for IoT: A Survey of Techniques, Challenges, and Applications

Journal of Sensor and Actuator Networks, Journal Year: 2025, Volume and Issue: 14(1), P. 9 - 9

Published: Jan. 22, 2025

Federated Learning (FL) has emerged as a pivotal approach for decentralized Machine (ML), addressing the unique demands of Internet Things (IoT) environments where data privacy, bandwidth constraints, and device heterogeneity are paramount. This survey provides comprehensive overview FL, focusing on its integration with IoT. We delve into motivations behind adopting FL IoT, underlying techniques that facilitate this integration, challenges posed by IoT environments, diverse range applications is making an impact. Finally, submission also outlines future research directions open issues, aiming to provide detailed roadmap advancing in settings.

Language: Английский

---

A Survey of Security Strategies in Federated Learning: Defending Models, Data, and Privacy

Future Internet, Journal Year: 2024, Volume and Issue: 16(10), P. 374 - 374

Published: Oct. 15, 2024

Federated Learning (FL) has emerged as a transformative paradigm in machine learning, enabling decentralized model training across multiple devices while preserving data privacy. However, the nature of FL introduces significant security challenges, making it vulnerable to various attacks targeting models, data, and This survey provides comprehensive overview defense strategies against these attacks, categorizing them into defenses privacy attacks. We explore pre-aggregation, in-aggregation, post-aggregation defenses, highlighting their methodologies effectiveness. Additionally, delves advanced techniques such homomorphic encryption differential safeguard sensitive information. The integration blockchain technology for enhancing environments is also discussed, along with incentive mechanisms promote active participation among clients. Through this detailed examination, aims inform guide future research developing robust frameworks systems.

Language: Английский

---

Blockchain-Based Decentralized Learning for Security in Digital Twins

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 10(24), P. 21479 - 21488

Published: July 14, 2023

This work aims to analyze malicious communication behaviors that pose a threat the security of digital twins (DTs) and safeguard user privacy. A unified integrated multidimensional DTs Network (DTN) architecture is constructed. On this basis, propagation process model malware in network built behavior threatens security. ensures protection mobile distributed machine learning system Blockchain technology data mechanism with broad prospects. It characterized by decentralization, transparency, anonymity, which can help ensure secure sharing privacy protection. Based on this, designs (DDS) based blockchain improve reliability support Internet Things (IoT). Then, resource allocation semi-distributed examined propose federated continuous (BL-FCL) algorithm combining DTs. significantly speeds up training process. Broad supports incremental learning. In way, each client does not need retrain when newly generated data. experimental part, prediction accuracy BL-FCL mixed national institute standards set similar FedAvg-50 FedAvg-80 schemes. As number devices increases from 1 6, detection probability exhibits rapid decrease. However, as further 6 10, gradually decreases at slower rate until it reaches 0. Comparatively, outperforms averaging algorithm-based scheme 20%–60%. The reported here deal problem inaccurate while ensuring users. great significance for DTN promoting development economy. results provide references applying DT field.

Language: Английский

---

Enhancing Security: Federated Learning against Man-In-The-Middle Threats with Gradient Boosting Machines and LSTM

Published: July 15, 2024

Language: Английский

---

Recent Advancements in Federated Learning: State of the Art, Fundamentals, Principles, IoT Applications and Future Trends

Future Internet, Journal Year: 2024, Volume and Issue: 16(11), P. 415 - 415

Published: Nov. 9, 2024

Federated learning (FL) is creating a paradigm shift in machine by directing the focus of model training to where data actually exist. Instead drawing all into central location, which raises concerns about privacy, costs, and delays, FL allows take place directly on device, keeping safe minimizing need for transfer. This approach especially important areas like healthcare, protecting patient privacy critical, industrial IoT settings, moving large numbers not practical. What makes even more compelling its ability reduce bias that can occur when are centralized, leading fairer inclusive outcomes. However, it without challenges—particularly with regard models secure from attacks. Nonetheless, potential benefits clear: lower costs associated storage processing, while also helping organizations meet strict regulations GDPR. As edge computing continues grow, FL’s decentralized could play key role shaping how we handle future, toward privacy-conscious world. study identifies ongoing challenges ensuring security against adversarial attacks, pointing further research this area.

Language: Английский

---

Resource-Efficient Clustered Federated Learning Framework for Industry 4.0 Edge Devices

AI, Journal Year: 2025, Volume and Issue: 6(2), P. 30 - 30

Published: Feb. 6, 2025

Industry 4.0 is an aggregate of recent technologies including artificial intelligence, big data, edge computing, and the Internet Things (IoT) to enhance efficiency real-time decision-making. data analytics demands a privacy-focused approach, federated learning offers viable solution for such scenarios. It allows each device train model locally using its own collected shares only updates with server without need share real data. However, communication computational costs sharing performance are major bottlenecks resource-constrained devices. This study introduces representative-based parameter-sharing framework that aims in environment. The begins by distributing initial devices, which then it send updated parameters back aggregation. To reduce costs, identifies groups devices similar parameter distributions sends from resourceful better-performing device, termed cluster head, server. A backup head also elected ensure reliability. Clustering performed based on device’s characteristics. Moreover, incorporates randomly selected past aggregated into current aggregation process through weighted averaging where more given greater weight performance. Comparative experimental evaluation state art testbed dataset demonstrates promising results minimizing cost while preserving prediction performance, ultimately enhances industrial environments.

Language: Английский

---

Guarding Privacy in Federated Learning: Exploring Threat Landscapes and Countermeasures with Case Studies

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 221 - 231

Published: Jan. 1, 2025

Language: Английский

---

Efficient clustered federated learning for Industrial Internet of Things: enhancing predictive performance and training time

Deleted Journal, Journal Year: 2025, Volume and Issue: 28(1)

Published: April 15, 2025

Abstract The Industrial Internet of Things (IIoT) brings together industrial devices in a network that gathers and analyzes data real-time for making data-driven decisions. Federated learning is popular approach collaboratively training multiple edge using an intermediate server rounds. This can be applied various fields, including anomaly detection, asset management, energy efficiency, quality control, predictive maintenance. However, performance affected by limited non-independent, identically distributed (non-IID) data. Additionally, also face resource constraints large datasets. paper proposes cluster-assisted custom federated improving the prediction resources required training. initializes model broadcasting initial parameters, then start After on current round’s data, transmit updated performance, distribution back to server. Then, clusters based their minimize non-IID. Parameter aggregation undertaken within cluster improve aggregated parameter sent respective members. Assuming secure internal network, work share samples round increase dataset size diversity. Earlier portion datasets are excluded from reduce drift. Comprehensive experimental evaluation with testbed proves effectiveness proposed over state-of-the-art.

Language: Английский

---

Advanced Machine Learning in Smart Grids: An Overview

Internet of Things and Cyber-Physical Systems, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

Language: Английский

---

IoT-Driven Waste Management in Smart Cities: Real-Time Monitoring and Optimization

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 413 - 425

Published: Jan. 1, 2025

Language: Английский

---