AI in Edge Computing for IoT Optimization

Relevance. The growing integration of Internet of Things (IoT) technologies into all areas of human life – from intelligent households to smart city infrastructure – is accompanied by an exponential increase in the volume of data being collected, transmitted, and processed in real time. When combined with artificial intelligence technologies, this data becomes the foundation for making autonomous decisions, predicting user behavior, and adapting environments to the needs of specific individuals. However, it is precisely in this context that the critically important issue of personal data protection arises. Many IoT devices operate in uncontrolled environments, have limited resources for cryptographic protection, and are vulnerable to cyberattacks and unauthorized data collection. Meanwhile, artificial intelligence algorithms used to analyze this data often exhibit the “black box” problem, where it is impossible to fully explain how and why a particular decision was made based on personalized data. The lack of transparency, combined with broad access to sensitive information, threatens fundamental human rights to privacy. The relevance of this topic is driven by the need to find balanced technical solutions that enable both effective analysis of large-scale data in IoT environments and a high level of data security. In this regard, the study of modern methods for processing, analyzing, and protecting data in IoT systems, adapted to the requirements of ethical artificial intelligence and digital privacy standards, represents one of the key challenges of contemporary digital science. The object of research: the processes of data collection, processing, analysis, and protection in IoT systems, particularly those components related to the use of users' personal information and its processing through artificial intelligence methods. Purpose of the article: research of modern methods for data processing and analysis in IoT systems. The objective of the work is to identify the most effective approaches to secure data handling, characterize existing privacy threats, and assess the potential for integrating protected analytical algorithms that meet both the technical and ethical requirements of the digital environment. Research results. A comprehensive analysis of modern approaches to data collection, processing, analysis, and protection in IoT systems has been conducted, particularly in the context of the growing role of artificial intelligence. The technological foundations of IoT functionality were examined, key architectural components identified, and their role in creating digital ecosystems for monitoring, management, and decision-making across various sectors – from household systems to critical infrastructure – was investigated. Special attention was paid to data preprocessing methods, which help reduce information load, improve the quality of analysis, and adapt data flows to the requirements of intelligent algorithms. It was demonstrated that the use of edge processing and local-level aggregation enhances both system performance and security. The main types of databases for IoT – especially those optimized for time series – were analyzed, along with tools for handling large volumes of data in cloud and hybrid environments. Conclusions. Data collection methods in IoT are multilayered and closely linked to the requirements for energy efficiency, security, latency, and system scalability. The quality and reliability of the collected information form the foundation for subsequent processing, analysis, and decision-making; therefore, the selection of sensors, communication protocols, and architectural models is of strategic importance for any IoT system. Preprocessing and efficient data storage in IoT are critical stages that ensure the quality, security, and usability of information for further analysis. They determine not only the accuracy of analytics but also the stability, scalability, and compliance with regulatory standards. This creates a demand for the development of adaptive, intelligent data processing and storage systems capable of dynamically responding to changes in device operation context and user requirements. The successful implementation of secure IoT solutions requires an integrated approach that combines technical expertise, legal knowledge, and ethical responsibility.

In recent years, significant advances have been made in health systems with the widespread adoption of the Internet of Things (IoT) and AI-powered innovative services. IoT enables the continuous collection of health data, while Artificial Intelligence (AI) analyzes this data to enhance early disease diagnosis. However, traditional Deep Learning (DL) approaches generally rely on centralized systems that collect and process personal data, posing a risk of violating individual privacy. In this study, a Federated Learning (FL) approach is used to process health data obtained from wearable smart devices while preserving privacy and classifying individuals' health scores. To overcome the performance limitations of FL approaches, a novel approach, FedSync, is proposed to incorporate a parallel training mechanism. To demonstrate the superiority of the FedSync approach, results are compared with DL and FL approaches in terms of accuracy, precision, recall, and F1-score. According to the results, the proposed approach provides a 29.02% improvement in computational time performance compared to FL and a 21.01% improvement compared to DL. Furthermore, to address the lack of explainability in IoT data, Explainable AI (XAI) has been integrated into the experiments to identify the most influential features affecting the classification score. This study makes a significant contribution to the literature by presenting a synchronous FL approach supported by XAI for IoT-based health data analysis.

Long-term, flexible therapy strategies are needed for chronic diseases like cardiovascular ailments, diabetes, and chronic kidney disease and they have to deal with patient heterogeneity, changing physiological states and privacy constraints. This research suggests a framework for privacy-preserving and personalized artificial intelligence that combines Federated Learning (FL), a Res-HyperTransformerNet deep prediction model, Personalized Multi-Agent Reinforcement Learning (PMARL) and Explainable Artificial Intelligence (XAI) for adaptive chronic disease management. The framework is structured with two main aims: (i) to create an accurate and privacy-aware predictive model for chronic disease risk using diverse data sources, and (ii) to permit adaptive, individual therapy optimization through multi-agent reinforcement learning. Federated learning is utilized to conduct the training of Res-HyperTransformerNet over distributed Internet of Medical Things (IoMT) nodes without the requirement of transferring raw patient data. The embeddings that are created from patient data are then passed on to a PMARL module, where several agents optimize therapy dimensions like medications, diet, physical activity, and mental health interventions independently. To make the clinical support more understandable, the SHAP-based explainability method is used for both predictive and decision-making parts. The framework is tested using two public datasets—the CDC Chronic Disease dataset and the UCI Chronic Kidney Disease Risk Factor dataset. The performance is measured by employing classification metrics (accuracy, precision, recall, F1-score, MCC) as well as reinforcement learning metrics (reward score, convergence steps, episode return) and federated system metrics (communication overhead, convergence rounds, and training time). The experiments indicate that the proposed framework demonstrated improved performance in terms of predictive accuracy and policy convergence speed compared to the baseline deep learning and reinforcement learning models, and at the same time it is more cost-effective when it comes to communication in a federated setting. This means that the proposed method suggests the potential of being a reproducible and open AI framework for adaptive chronic disease therapy management. The combination of ResNet and Transformer blocks, i.e., Federated Res-HyperTransformerNet, achieved strong performance in both datasets, namely CDC Chronic Disease (Dataset 1) and UCI CKD Risk Factor (Dataset 2). The obtained accuracy for Dataset 1 was 98.61% and for Dataset 2, it was 97.75%, under the conducted experimental settings.

Artificial intelligence (AI) has, in the recent past, experienced a rebirth with the growth of generative AI systems such as ChatGPT and Bard. These systems are trained with billions of parameters and have enabled widespread accessibility and understanding of AI among different user groups. Widespread adoption of AI has led to the need for understanding how machine learning (ML) models operate to build trust in them. An understanding of how these models generate their results remains a huge challenge that explainable AI seeks to solve. Federated learning (FL) grew out of the need to have privacy-preserving AI by having ML models that are decentralized but still share model parameters with a global model. This study sought to examine the extent of development of the explainable AI field within the FL environment in relation to the main contributions made, the types of FL, the sectors it is applied to, the models used, the methods applied by each study, and the databases from which sources are obtained. A systematic search in 8 electronic databases, namely, Web of Science Core Collection, Scopus, PubMed, ACM Digital Library, IEEE Xplore, Mendeley, BASE, and Google Scholar, was undertaken. A review of 26 studies revealed that research on explainable FL is steadily growing despite being concentrated in Europe and Asia. The key determinants of FL use were data privacy and limited training data. Horizontal FL remains the preferred approach for federated ML, whereas post hoc explainability techniques were preferred. There is potential for development of novel approaches and improvement of existing approaches in the explainable FL field, especially for critical areas. OSF Registries 10.17605/OSF.IO/Y85WA; https://osf.io/y85wa.

Background The growing scale and complexity of Internet of Things (IoT) environments have intensified the need for intelligent and adaptive cybersecurity mechanisms. Artificial intelligence (AI)-based intrusion detection systems (IDS) have emerged as a promising solution for identifying and mitigating threats in real time. Methodology This review systematically evaluates the effectiveness of AI-based IDS in IoT networks, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. A comprehensive search of the Scopus and Web of Science databases was conducted, yielding 203 studies, of which 51 met the inclusion criteria. Eligible studies, published between 2016 and 2025, were analyzed for geographic distribution, AI techniques used, methodological quality, and reported outcomes. Meta-regression and contour-enhanced funnel plots were employed to assess effect size trends and publication bias. Results Most studies originated from India, Saudi Arabia, and China, with research output peaking in 2024. Meta-regression analysis revealed a positive correlation between publication year and reported effect size, indicating progressive advancements in AI methodologies. Machine learning (ML) and deep learning (DL) were the most widely used techniques, with a growing trend toward hybrid and ensemble models that enhance threat detection accuracy. Recent studies also showed increased interest in explainable artificial intelligence (XAI), reflecting the demand for transparency and interpretability in model outputs. Funnel plot asymmetry suggested a bias toward publishing positive findings. Conclusions AI-based IDSs have demonstrated substantial potential in strengthening IoT security, particularly through ML, DL, and hybrid approaches. However, inconsistencies in evaluation metrics, reporting standards, and methodological design remain significant challenges. The findings highlight the need for standardized benchmarks and robust frameworks to guide future research and ensure reliable deployment of AI-driven IDS in diverse IoT contexts.

With the continuous development of the Internet of Things (IoT) and the needs of the times, the number of IoT edge devices is rapidly increasing, and to be able to provide users with more reliable information services and decision-making, this paper constructs an IoT system based on cloud-edge collaboration. Based on cloud-edge collaboration, data mining technology is used to categorize data, and regression models and branch neural networks are combined to further improve the IoT system based on cloud-edge collaboration. Finally, the particle swarm algorithm is used to optimize reinforcement learning, and the framework of an IoT system built on cloud-edge partnership is established. Through the data processing experiments on IoT edge computing, the data mining performance of the IoT edge computing-based data mining method is improved by 0.0295, 0.1870, and 0.2059 on the large datasets of Dataset4 and Dataset5, respectively. the average turnaround time is lower than 6.5 for different edge computing power levels. The system rewards are gradually improved from 5.3 to 8.2, and the number of packet loss of the algorithm is less than 5. The feasibility and effectiveness of the IoT data mining method and system based on cloud-edge collaboration are confirmed by this study, and a feasible method for data mining can be applied to IoT systems under cloud-edge collaboration.

Enhanced Dynamic Deep Q-Network for Federated Learning scheduling policies on IoT devices using explanation-driven trust

The integration of Internet of Things (IoT) and Edge Artificial Intelligence (AI) has revolutionized precision agriculture by providing real-time, data-driven insights for optimizing farming practices. As IoT devices proliferate in agricultural environments, the need for robust and scalable systems that can seamlessly handle vast amounts of data becomes critical. This chapter explores the challenges and opportunities presented by large-scale IoT deployments in agriculture, focusing on the role of Edge AI in ensuring real-time processing and decision-making. Key issues such as device heterogeneity, scalability, interoperability, and data security are examined, alongside strategies for effective system integration and validation. The chapter delves into case studies that illustrate the successful deployment of IoT and Edge AI technologies in diverse agricultural settings, highlighting lessons learned and best practices. Furthermore, it outlines future directions for enhancing system efficiency, ensuring secure data transmission, and fostering widespread adoption of these technologies in precision farming. By addressing these challenges, the chapter provides a comprehensive framework for developing and deploying scalable IoT and Edge AI systems that can drive sustainable agricultural practices worldwide.

Recent years have witnessed the widespread popularity of Internet of things (IoT). By providing sufficient data for model training and inference, IoT has promoted the development of artificial intelligence (AI) to a great extent. Under this background and trend, the traditional cloud computing model may nevertheless encounter many problems in independently tackling the massive data generated by IoT and meeting corresponding practical needs. In response, a new computing model called edge computing (EC) has drawn extensive attention from both industry and academia. With the continuous deepening of the research on EC, however, scholars have found that traditional (non-AI) methods have their limitations in enhancing the performance of EC. Seeing the successful application of AI in various fields, EC researchers start to set their sights on AI, especially from a perspective of machine learning, a branch of AI that has gained increased popularity in the past decades. In this article, we first explain the formal definition of EC and the reasons why EC has become a favorable computing model. Then, we discuss the problems of interest in EC. We summarize the traditional solutions and hightlight their limitations. By explaining the research results of using AI to optimize EC and applying AI to other fields under the EC architecture, this article can serve as a guide to explore new research ideas in these two aspects while enjoying the mutually beneficial relationship between AI and EC.

Usually, IoT devices are designed to perform specific tasks, while robots have to adapt to unpredictable situations. Artificial intelligence and machine learning help these robots cope with the emerging unexpected conditions. The Internet of Robotic Things is an evolving concept that brings together all-encompassing sensors and devices with robotic and autonomous systems. Both IoT devices and robots rely on sensors to understand the surrounding environment, to process data quickly, and to decide how to respond. Nevertheless, while most IoT systems can handle only well-defined tasks, robots can handle expected situations as well. The Internet of Robotic Things is a better Internet of Things (IoT) solution due to its ability to bridge the gap between IT and real operations. The IoRT may face several challenges in power consumption and network bandwidth limitation due to the growth in the information dissemination over the Robotic network. The main objective of this paper is to introduce an overview of the concepts and challenges in the Internet of Robotic Things based on Fog Computing technique. Further, a framework of Fog based IoRT is introduced.

In the 21st century, Urbanization, population growth, and climate change have created significant problems in water resource management. Recent advancements in technologies such as Internet of Things (IoT), Edge Computing (EC), Artificial Intelligence (AI), and Big Data Analytics (BDA) are changing the operations of the water resource management systems. In this study, we present a systematic review, highlighting the contributions of these technologies in water management systems. More specifically, we highlight the IoT and EC water monitoring systems that enable real-time sensing of water quality and consumption. In addition, AI methods for anomaly detection and predictive maintenance are reviewed, focusing on water demand forecasting. BDA methods are also discussed, highlighting their ability to integrate data from different data sources, such as sensors and historical data. Additionally, a discussion is provided of how Water management systems could enhance sustainability, resilience, and efficiency by combining big data, IoT, EC, and AI. Lastly, future directions are outlined regarding how state-of-the-art technologies may further support efficient water resources management.

Commonly, an Internet of Things (IoT) system follows the architecture of the Cloud‐centric Internet of Things (CIoT) in which the physical objects are represented in the form of Web resources that are managed by the servers in the global Internet. In general, the common IoT system involves three major technologies: embedded systems, middleware, and cloud services. Although the CIoT model is a common approach to implement IoT systems, it is facing the growing challenges in IoT. This chapter discusses foundations of computing paradigms for realizing emerging IoT applications, especially fog and edge computing, their background, characteristics, architectures and open challenges. It presents related technologies to fog and edge computing. The chapter describes how fog and edge can improve CIoT and explains the hierarchy of fog and edge computing environments. It illustrates the business models of fog and edge computing, and provides the information regarding the opportunities and challenges in fog and edge computing.

In recent years, the government, universities, and industry have paid attention to the distributed approach in machine learning (ML) and cybersecurity for the developing Internet of Things (IoT). Federated cybersecurity (FC) is a strategy for making the Internet of Things (IoT); more secured and effective in the future. This new approach has an ability to detect problems regarding security, implement solutions, and across the IoT systems it manages effectively to limit the propagation of vulnerabilities. Forming a federation of shared information is one way to attain a cybersecurity. Federated learning (FL) is a machine learning paradigm which is especially effective for securing the sensitive IoT environment. The origin of FL, and also FL for cyber security are presented in this article. The various security assaults, and responses are also discussed as an outcome of this article. Experiments are carried out in Google Colaboratory, using well-known python libraries. The results shows that FL provides highest level of security in comparison with centralised learning.

The Internet of Things (IoT) is penetrating various facets of our daily life with the proliferation of intelligent services and applications empowered by artificial intelligence (AI). AI techniques require centralized data collection and processing that may not be feasible in realistic application scenarios due to the high scalability of modern IoT networks and growing data privacy concerns. Federated Learning (FL) has emerged as a distributed collaborative AI approach that can enable many intelligent IoT applications by allowing for AI training at distributed IoT devices without the need for data sharing. In this survey, we focus on DT and FL for IIoT. Initially, we analyzed the existing surveys. In this paper, we present the applications of DT and FL in IIoT.

The cutting edge of Industry 4.0 has driven everything to be converted to disruptive innovation and digitalized. This digital revolution is imprinted by modern and advanced technology that takes advantage of Big Data and Artificial Intelligence (AI) to nurture from automatic learning systems, smart city, smart energy, smart factory to the edge computing technology, and so on. To harness an appealing, noteworthy, and leading development in smart manufacturing industry, the modern industrial sciences and technologies such as Big Data, Artificial Intelligence, Internet of things, and Edge Computing have to be integrated cooperatively. Accordingly, a suggestion on the integration is presented in this paper. This proposed paper describes the design and implementation of big data platform for intelligence industrial internet of things sensor monitoring system and conveys a prediction of any upcoming errors beforehand. The architecture design is based on edge computing and artificial intelligence. To extend more precisely, industrial internet of things sensor here is about the condition monitoring sensor data - vibration, temperature, related humidity, and barometric pressure inside facility manufacturing factory.