Payload State Prediction Based on Real-Time IoT Network Traffic Using Hierarchical Clustering with Iterative Optimization DOI Creative Commons
Hao Zhang, Wang Jing, Xuanyuan Wang

и другие.

Sensors, Год журнала: 2024, Номер 25(1), С. 73 - 73

Опубликована: Дек. 26, 2024

IoT (Internet of Things) networks are vulnerable to network viruses and botnets, while facing serious security issues. The prediction payload states in can detect attacks achieve early warning rapid response prevent potential threats. Due the instability packet loss communications between victim nodes, constructed protocol state machines existing schemes inaccurate. In this paper, we propose a predictor called IoTGuard, which predict based on real-time traffic. steps IoTGuard briefly as follows: Firstly, application-layer payloads different nodes extracted through module separation. Secondly, classification within flows is obtained via extraction module. Finally, trained set, these have labels. Experimental results Mozi botnet dataset show that more accurately ensuring execution efficiency. achieves an accuracy 86% prediction, 8% higher than state-of-the-art method NetZob, training time reduced by 52.8%.

Язык: Английский

A multi objective optimization framework for smart parking using digital twin pareto front MDP and PSO for smart cities DOI Creative Commons
Dinesh Kumar Sahu, Pradip Sinha, Shiv Prakash

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 5, 2025

Smart cities are designed to improve the quality of life by efficiently using resources and smart parking is an important part this puzzle help alleviate traffic congestion address energy consumption search time for spaces. However, existing management systems have issues with resource management, system scalability, real-time dynamic changes. In response these challenges, paper proposes a Multi-Objective Optimization Framework Parking incorporating Digital Twin Technology, Pareto Front Optimization, Markov Decision Process (MDP), Particle Swarm (PSO). Hence, proposed framework utilizes whereby there generation virtual model infrastructure that can give prospective estimation entire system. The then used multi-objective optimization domain, where goal minimize time, use energy, disruption, maximize availability MDP splits allocation problem into value function which requests. Further, PSO refines solutions found from front globally superior distribution. evaluated extensive simulations across multiple metrics: level, utilization. Evaluation outcomes also show algorithm better than Round Robin, Random Allocation, Threshold Based algorithms in terms 25% improvement 18% usage, 30% less congestion. This work has shown prospects combining hybrid decision-making enhancement efficiency urban mobility.

Язык: Английский

Процитировано

1
Federated Learning-Based Predictive Traffic Management Using a Contained Privacy-Preserving Scheme for Autonomous Vehicles DOI Creative Commons
Tariq Alqubaysi, Abdullah Faiz Al Asmari, Fayez Alanazi

и другие.

Sensors, Год журнала: 2025, Номер 25(4), С. 1116 - 1116

Опубликована: Фев. 12, 2025

Intelligent Transport Systems (ITSs) are essential for secure and privacy-preserving communications in Autonomous Vehicles (AVs) enhance facilities like connectivity roadside assistance. Earlier research models used traffic management compromised user privacy exposed sensitive data to potential adversaries while handling real-time from numerous vehicles. This introduces a Federated Learning-based Predictive Traffic Management (FLPTM) system designed optimize service access within an ITS. Moreover, CPPS will provide strong security mitigate adversarial threats through state modelling authenticated permissions the integrity of vehicle communication networks man-in-the-middle attacks. The suggested FLPTM utilizes Contained Privacy-Preserving Scheme (CPPS) that decentralizes processing allows vehicles train local without sharing raw data. framework combines classifier-based learning technique with protect against invasions proposed model leverages Learning (FL) collaborative machine processes by allowing updates preserve privacy, enabling joint exposing It addresses key challenges such as high costs, impact attacks, time inefficiencies. Using FL, reduces costs 23.29%, mitigates effects 16.1%, improves 18.95%, achieving significant cost savings maintaining throughout process.

Язык: Английский

Процитировано

1
AI-Driven UAV and IoT Traffic Optimization: Large Language Models for Congestion and Emission Reduction in Smart Cities DOI Creative Commons

Álvaro Moraga,

J. de Curtò, I. de Zarzà

и другие.

Drones, Год журнала: 2025, Номер 9(4), С. 248 - 248

Опубликована: Март 26, 2025

Traffic congestion and carbon emissions remain pressing challenges in urban mobility. This study explores the integration of UAV (drone)-based monitoring systems IoT sensors, modeled as induction loops, with Large Language Models (LLMs) to optimize traffic flow. Using SUMO simulator, we conducted experiments three scenarios: Pacific Beach Coronado San Diego, Argüelles Madrid. A Gemini-2.0-Flash experimental LLM was interfaced simulation dynamically adjust vehicle speeds based on real-time conditions. Comparative results indicate that AI-assisted approach significantly reduces CO2 compared a baseline without AI intervention. research highlights potential UAV-enhanced frameworks for adaptive, scalable management, aligning future drone-assisted mobility solutions.

Язык: Английский

Процитировано

1
Distributed computing in multi-agent systems: a survey of decentralized machine learning approaches DOI
Ijaz Ahmed, Miswar Akhtar Syed, Muhammad Maaruf

и другие.

Computing, Год журнала: 2024, Номер 107(1)

Опубликована: Ноя. 19, 2024

Язык: Английский

Процитировано

3
Blockchain Technology and Smart Cities: A Technological Framework for Innovation and Sustainability in the UAE and Beyond DOI Creative Commons
Saeed Ali Faris Alketbi, Massudi Mahmuddin, Mazida Ahmad

и другие.

Data & Metadata, Год журнала: 2025, Номер 4, С. 697 - 697

Опубликована: Фев. 11, 2025

Introduction: Blockchain technology has emerged as a cornerstone for innovation in the field of information systems, offering secure, decentralized, and transparent solutions to address complex challenges smart city development. This paper explores transformative potential blockchain advancing cities, focusing on its ability integrate with Internet Things (IoT) enable secure data management, optimize urban services. Key challenges, such scalability, interoperability, regulatory frameworks, are analyzed alongside innovative solutions, including second-layer protocols, cross-chain communication, energy-efficient consensus mechanisms. The study introduces International Certification Layer (ICL) novel framework designed enhance oversight while maintaining blockchain’s decentralized integrity. Additionally, Dubai’s Strategy serves pioneering case study, showcasing how strategic investment can streamline governance, citizen trust, support achievement sustainability goals. By addressing critical identifying future research directions, this underscores role enabler sustainable efficient ecosystems.

Язык: Английский

Процитировано

0
Enhancing healthcare data privacy and interoperability with federated learning DOI Creative Commons

Adil Akhmetov,

Zohaib Latif,

Benjamin Tyler

и другие.

PeerJ Computer Science, Год журнала: 2025, Номер 11, С. e2870 - e2870

Опубликована: Май 8, 2025

This article explores the application of federated learning (FL) with Fast Healthcare Interoperability Resources (FHIR) protocol to address underutilization huge volumes healthcare data generated by digital health revolution, especially those from wearable sensors, due privacy concerns and interoperability challenges. Despite advances in electronic medical records, mobile applications, current cannot fully exploit these lack analysis exchange between heterogeneous systems. To this gap, we present a novel converged platform combining FL FHIR, which enables collaborative model training that preserves sensor while promoting standardization interoperability. Unlike traditional centralized (CL) solutions require centralization, our uses local learning, naturally improves privacy. Our empirical evaluation demonstrates models perform as well as, or even numerically better than, terms classification accuracy, also performing equally regression, indicated metrics such area under curve (AUC), recall, precision, among others, for classification, mean absolute error (MAE), squared (MSE), root square (RMSE) regression. In addition, developed an intuitive AutoML-powered web is CL compatible illustrate feasibility predictive modeling physical activity energy expenditure, complying FHIR reporting standards. These results highlight immense potential FHIR-integrated practical framework future interoperable privacy-preserving ecosystems optimize use connected data.

Язык: Английский

Процитировано

0
AI-Powered Hybrid Smart Parking: Optimizing Parking Management Across Diverse Applications in Smart Cities DOI Open Access
S. K.,

J. Hanumanthappa,

Kanhaiya Kumar

и другие.

Procedia Computer Science, Год журнала: 2025, Номер 258, С. 1524 - 1535

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Empowering Smart Cities: Unlocking Citizen Participation Through AI-Driven Personalization and Perceived Value DOI Creative Commons

Afshar Hatami,

Somayyeh NaserAmini Jeloudarlou,

Haniyeh Asadzadeh

и другие.

Sustainable Futures, Год журнала: 2025, Номер unknown, С. 100664 - 100664

Опубликована: Май 1, 2025

Язык: Английский

Процитировано

0
A Comprehensive Survey on the Societal Aspects of Smart Cities DOI Creative Commons
David Bastos, Nuno Costa, Nelson Pacheco Rocha

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(17), С. 7823 - 7823

Опубликована: Сен. 3, 2024

Smart cities and information communications technology is a rapidly growing field in both research real-world implementation, but it one that still new with many different ideas. Unfortunately, there less cooperation knowledge sharing across the field, often fails to move into applications, which holds back from becoming fully realized. This paper aims provide an overview of current state smart cities, its definitions, technologies, technical dimensions, architectural design standards data handling, how they are handled real world impact on society. Additionally, examines important city projects, their ranking systems. text forecast future impact, challenges faces, what should be addressed help reach full potential.

Язык: Английский

Процитировано

2
Payload State Prediction Based on Real-Time IoT Network Traffic Using Hierarchical Clustering with Iterative Optimization DOI Creative Commons
Hao Zhang, Wang Jing, Xuanyuan Wang

и другие.

Sensors, Год журнала: 2024, Номер 25(1), С. 73 - 73

Опубликована: Дек. 26, 2024

IoT (Internet of Things) networks are vulnerable to network viruses and botnets, while facing serious security issues. The prediction payload states in can detect attacks achieve early warning rapid response prevent potential threats. Due the instability packet loss communications between victim nodes, constructed protocol state machines existing schemes inaccurate. In this paper, we propose a predictor called IoTGuard, which predict based on real-time traffic. steps IoTGuard briefly as follows: Firstly, application-layer payloads different nodes extracted through module separation. Secondly, classification within flows is obtained via extraction module. Finally, trained set, these have labels. Experimental results Mozi botnet dataset show that more accurately ensuring execution efficiency. achieves an accuracy 86% prediction, 8% higher than state-of-the-art method NetZob, training time reduced by 52.8%.

Язык: Английский

Процитировано

0