The Internet of Medical Things (IoMT) has a bright future with the development smart mobile devices. Information technology is also leading changes in healthcare industry. IoMT devices can detect patient signs and provide treatment guidance even instant diagnoses through technologies, such as artificial intelligence (AI) wireless communication. However, conventional centralized machine learning approaches are often difficult to apply within because difficulty large-scale collection data...

Although AI-empowered schemes bring some sound solutions to stimulate more reasonable energy distribution between charging stations (CSs) and CS providers, frequent data sharing them is possible incur many security privacy breaches. To solve these problems, federated learning (FL) an ideal solution that only requires CSs upload local models instead of detailed data. the CSs' electricity consumption need not be exposed server directly, FL-based still have been excavated several threats such...

The rapid development of Internet Things (IoT) stimulates the innovation for health-related devices such as remote patient monitoring, connected inhalers and ingestible sensors. Simultaneously, with aid numerous equipments, a great number collected data can be used disease prediction or diagnosis model establishment. However, potential leak will also bring privacy security issues in interaction period. To deal these existing issues, we propose decentralized, efficient, privacy-enhanced...

The rapid growth of artificial intelligence (AI), blockchain technology, and edge computing services have enabled the Internet Medical Things (IoMT) to provide various healthcare patients, including neural network-based disease diagnosis, heart rate monitoring, fall detection. Generally, end devices should transmit collected patient data a centralized server for further model training, but at same time, patient's privacy may be risk. In addition, due diversity conditions, one-size-fits-all...

Federated learning can collaboratively train a global model without gathering clients' private data. Many works focus on reducing communication cost by designing kinds of client selection method or averaging algorithm. But they all consider whether the will participant not, and training time could not be reduced as data size update for each is changed. We proposed COFEL, novel federated system which both reduce layer-based parameter enhance privacy protection applying local differential...

Abstract Central management of electronic medical systems faces a major challenge because it requires trust in single entity that cannot effectively protect files from unauthorized access or attacks. This makes difficult to provide some services central systems, such as file search and verification, although they are needed. gap motivated us develop system based on blockchain has several characteristics: decentralization, security, anonymity, immutability, tamper-proof. The proposed provides...

In recent years, federated learning has attracted more and attention as it could collaboratively train a global model without gathering the users' raw data. It brought many challenges. this paper, we proposed layer-based system with privacy preservation. We successfully reduced communication cost by selecting several layers of to upload for averaging enhanced protection applying local differential privacy. evaluated our in non independently identically distributed scenario on three datasets....

With the improvement of computing power and development network technology, Internet Things (IoT) devices are widely used in many industries. But it also faces various security threats. Anomaly detection is a commonly method, but traditional methods face shortcomings such as low accuracy. Therefore, this paper, we introduce decentralized federated learning method for anomaly detection, using neural networks to improve accuracy take advantage characteristics protect local data security. The...

In the era of Internet Things (IoT) and AI-driven smart environments, human behavior recognition has emerged as a pivotal technology underpinning broad spectrum intelligent applications. However, achieving high accuracy while preserving user privacy remains critical challenge. To tackle this problem, paper introduces novel privacy-preserving method for WiFi-based recognition. It employs three-dimensional convolutional neural networks(3D-CNNs) enhanced with an attention-enabled autoencoder...

Traffic sign recognition (TSR) is vital for vehicle safety and navigation, especially in the era of autonomous cars. Internet Vehicles (IoV) provide a promising infrastructure vehicular networks due to their agility interoperability. However, privacy concerns network restrictions hinder collection massive data from distributed automotive sensors IoV. To address these challenges, this article proposes application federated learning (FL) model sparsification optimize traffic vehicles. FL...

Abstract Federated generative adversarial networks are designed to collaborate across the communication and privacy-constrained edge servers participating in training. However, Internet of Things scenario, local updates uploaded by can lead risk privacy breaches. Gradient-sanitized-based approaches transmit sanitized sensitive data with strict guarantees, but gradient clipping perturbation severely degrade convergence performance. In this paper, our proposed algorithm enhances terminated raw...

Device-free localization technology aims to find a target by analyzing the signal strength difference between transmitter and receiver deployed in area advance. Up now, device-free has been applied wide range of applications scenarios, such as intrusion detection, environment modeling, activity recognition. However, some sensors remain at potential risk that values have tampered, or even devices are physically damaged, which leads inaccurate location results whole system crash. To solve...

The Internet of Things (IoT) has evolved into a global platform dramatically facilitating human life through intelligent services. It is straightforward for people to access smart devices IoT. However, the easy accessibility IoT also led unprecedented security challenges To ensure basic structure IoT, we need establish barrier that can filter malicious and achieve integration intrusion detection systems (IDS) with gateways. This paper establishes threat models DoS, Replay, MITM, Loophole...

For data isolated islands and privacy issues, federated learning has been extensively invoking much interest since it allows clients to collaborate on training a global model using their local without sharing any with third party. However, the existing frameworks always need sophisticated condition configurations (e.g., driver configuration of standalone graphics card like NVIDIA, compile environment) that bring inconvenience for large-scale development deployment. To facilitate deployment...

The development and popularity of social networks have made information dissemination unprecedentedly convenient speedy. However, the spread fake news can often cause serious harm to society individuals. Therefore, machine learning-based detection methods become increasingly important. existing work needs collect sufficient user-side data for training, which also boosts privacy leakage risk users. this article proposes an intelligent system based on federated learning (FL) called FIND, train...