nach oben

2024 | Buch

Learning Techniques for the Internet of Things

herausgegeben von: Praveen Kumar Donta, Abhishek Hazra, Lauri Lovén

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

The book is structured into thirteen chapters; each comes with its own dedicated contributions and future research directions. Chapter 1 introduces IoT and the use of Edge computing, particularly cloud computing, and mobile edge computing. This chapter also mentions the use of edge computing in various real-time applications such as healthcare, manufacturing, agriculture, and transportation. Chapter 2 motivates mathematical modeling for federated learning systems with respect to IoT and its applications. Further Chapter 3 extends the discussion of federated learning for IoT, which has emerged as a privacy-preserving distributed machine learning approach. Chapter 4 provides various machine learning techniques in Industrial IoT to deliver rapid and accurate data analysis, essential for enhancing production quality, sustainability, and safety. Chapter discusses the potential role of data-driven technologies, such as Artificial Intelligence, Machine Learning, and Deep Learning, focuses on their integration with IoT communication technologies. Chapter 6 presents the requirements and challenges to realize IoT deployments in smart cities, including sensing infrastructure, Artificial Intelligence, computing platforms, and enabling communications technologies such as 5G networks. To highlight these challenges in practice, the chapter also presents a real-world case study of a city-scale deployment of IoT air quality monitoring within Helsinki city. Chapter 7 uses digital twins within smart cities to enhance economic progress and facilitate prompt decision-making regarding situational awareness. Chapter 8 provides insights into using Multi-Objective reinforcement learning in future IoT networks, especially for an efficient decision-making system. Chapter 9 offers a comprehensive review of intelligent inference approaches, with a specific emphasis on reducing inference time and minimizing transmitted bandwidth between IoT devices and the cloud. Chapter 10 summarizes the applications of deep learning models in various IoT fields. This chapter also presents an in-depth study of these techniques to examine new horizons of applications of deep learning models in different areas of IoT. Chapter 11 explores the integration of Quantum Key Distribution (QKD) into IoT systems. It delves into the potential benefits, challenges, and practical considerations of incorporating QKD into IoT networks. In chapter 12, a comprehensive overview regarding the current state of quantum IoT in the context of smart healthcare is presented, along with its applications, benefits, challenges, and prospects for the future. Chapter 13 proposes a blockchain-based architecture for securing and managing IoT data in intelligent transport systems, offering advantages like immutability, decentralization, and enhanced security.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Edge Computing for IoT

Abstract

Over the past few years, the idea of edge computing has seen substantial expansion in both academic and industrial circles. This computing approach has garnered attention due to its integrating role in advancing various state-of-the-art technologies such as Internet of Things (IoT), 5G, artificial intelligence, and augmented reality. In this chapter, we introduce computing paradigms for IoT, offering an overview of the current cutting-edge computing approaches that can be used with IoT. Furthermore, we go deeper into edge computing paradigms, specifically focusing on cloudlet and mobile edge computing. After that, we investigate the architecture of edge computing-based IoT, its advantages, and the technologies that make edge computing-based IoT possible, including artificial intelligence and lightweight virtualization. Additionally, we review real-life case studies of how edge computing is applied in IoT-based intelligent systems, including areas like healthcare, manufacturing, agriculture, and transportation. Finally, we discuss current research obstacles and outline potential future directions for further investigation in this domain.

Balqees Talal Hasan, Ali Kadhum Idrees

Chapter 2. Federated Learning Systems: Mathematical Modeling and Internet of Things

Abstract

Since the creation of the first computer in 1948, notably thanks to the work of Alan Turing with the universal machine in 1936, computing has become an integral part of our lives and occupies an important place in our society. Today, it would be hard to imagine a world without computers and mobile phones. All these tools have made it possible to develop new activities and also to increase our knowledge. Indeed, the calculation units are more and more powerful, and today it is possible to solve complex problems. It is in this logic of learning that machine learning was born in the late 1950s, with the creation of a program by Arthur Samuel, an American computer scientist who pioneered the development of artificial intelligence. This program, created for the giant IBM, played checkers by improving with each turn. Machine learning is a branch of artificial intelligence that aims to develop algorithms and models that can learn from data and make decisions or make predictions without being programmed. It is a booming technology that could revolutionize many areas of today’s society. The continuous development of machine learning is crucial to address the complex challenges facing our society and to drive innovation in various fields. By investing in the necessary research, education, and resources, we can unlock the full potential of machine learning and create a future where decisions are informed by accurate predictive models and repetitive tasks are automated, freeing up time. The latter can be used for activities with higher added value. In this chapter, we investigate a modern development in learning systems with the creation of federated learning (FL) and its application to Internet of Things (IoT) domains. Federated learning forces some computation to devices themselves, thereby limiting the amount of data that needs to be transmitted to central servers for computation.

Quentin De La Cruz, Gautam Srivastava

Chapter 3. Federated Learning for Internet of Things

Abstract

The proliferation of the Internet of Things (IoT) and the advancements in machine learning (ML) have facilitated ubiquitous sensing and computing capabilities, enabling the interconnection of a wide array of devices to the Internet. Traditionally, data collection and data processing have been centralized, which may not be feasible due to issues such as long propagation delays, communication overload, and increasing data privacy concerns. To tackle these challenges, federated learning (FL) has emerged as a privacy-preserving distributed ML approach, allowing numerous devices to engage in model training without transferring their local data to a central server. This work presents a comprehensive review of FL as an approach to performing ML on distributed IoT data, with a specific emphasis on protecting data privacy and reducing communication costs associated with data transfer. The review encompasses various aspects, including the background of FL, the architecture of FL for IoT, the different types of FL for IoT, FL frameworks tailored for IoT, and diverse FL for IoT applications. Additionally, this paper outlines future research challenges and directions pertaining to FL for IoT. By embracing the potential of FL while addressing its challenges, IoT can benefit from reduced delays, improved communication efficiency, enhanced privacy preservation, and a more sustainable FL-IoT system.

Ying Li, Qiyang Zhang, Xingwei Wang, Rongfei Zeng, Haodong Li, Ilir Murturi, Schahram Dustdar, Min Huang

Chapter 4. Machine Learning Techniques for Industrial Internet of Things

Abstract

Industrial Internet of Things (IIoT), which connects millions of smart devices, will allow for industrial use cases like smart cities and supply chain management with minimal human involvement in the future. The IIoT has revolutionized production by making data faster, more accurate, and more accessible to stakeholders at all levels. In the IIoT, machine learning (ML) techniques are frequently utilized to add intelligence to the industrial environment and manufacturing operations. For instance, timely and accurate data analysis is essential, and ML techniques are used to examine and comprehend the enormous amounts of data created by IoT devices. Organizations use ML algorithms to promote innovation, make smart decisions, and create autonomous industrial environments. IoT and ML are employed in manufacturing to enhance quality control, streamline production, and cut waste. For instance, producers can spot areas for improvement and carry out preventative maintenance before equipment faults occur by applying ML algorithms to analyze data from IoT sensors on factory equipment. Learning techniques in IIoT are critical to deliver rapid and accurate data analysis, essential for enhancing production quality, sustainability, and safety. Motivated by the abovementioned learning technology, in this chapter, we discuss the significance of ML and its benefits toward IIoT for processing real-time applications. We shed light on several key ML technologies for IIoT. Finally, we highlight several research challenges and outstanding concerns that need further addressing to realize the IIoT scenario.

Megha Sharma, Abhishek Hazra, Abhinav Tomar

Chapter 5. Exploring IoT Communication Technologies and Data-Driven Solutions

Abstract

Over the past decade, Internet of Things (IoT) networks have been the subject of active research due to their wide range of potential applications. The successful implementation and effective performance of IoT networks depend on the communication protocols used to connect spatially distributed devices or sensors. However, existing communication technologies face several challenges, including security, interoperability, scalability, and energy optimization. Therefore, researchers are currently exploring novel IoT communication protocols and embracing data-driven approaches along with other solutions to overcome these challenges. This chapter comprehensively explores emerging trends in IoT communication technologies and the integration of data-driven solutions. Additionally, we study the potential role of data-driven technologies, such as artificial intelligence (AI), machine learning (ML), and deep learning (DL), focusing on their integration with IoT technologies. We have also briefly discussed the benefits of using data-driven technologies in various IoT applications. Furthermore, we have outlined several potential challenges and how data-driven technologies can address them, emphasizing recent innovations.

Poonam Maurya, Abhishek Hazra, Lalit Kumar Awasthi

Chapter 6. Towards Large-Scale IoT Deployments in Smart Cities: Requirements and Challenges

Abstract

The Internet of Things (IoT) plays a significant role in the development and future evolution of smart cities by connecting physical devices and systems to the Internet to collect and exchange data, automate processes, and improve overall urban management and quality of life. This chapter presents the requirements and challenges to realize IoT deployments in smart cities, including sensing infrastructure, artificial intelligence (AI), computing platforms, and enabling communication technologies such as 5G beyond networks. To highlight these challenges in practice, the chapter also presents a real-world case study of a city-scale deployment of IoT air quality monitoring within the city of Helsinki. The results demonstrate the role that IoT plays in future smart cities, illustrating how deployments of air quality monitoring devices can benefit decision-making by supporting local air pollution monitoring, traffic management, and urban planning. Lastly, the chapter discusses the role of AI and other emerging technologies on the future of smart cities.

Naser Hossein Motlagh, Martha Arbayani Zaidan, Roberto Morabito, Petteri Nurmi, Sasu Tarkoma

Chapter 7. Digital Twin and IoT for Smart City Monitoring

Abstract

This study involves the integration of wireless technologies to enable the creation of digital twin representations for the purpose of monitoring smart cities. The analytical representations are created through the utilization of advanced technological resources, facilitating a comparison between the original values and reference values. Consequently, the time step index is accompanied by error measurements that have been reduced, in cases where a greater number of active messages are present, along with the creation of twins that are transmitted in a secure manner. The primary objective of digital twins within the context of smart cities is to enhance economic progress and facilitate prompt decision-making with respect to situational awareness. To this end, the Internet of Things (IoT) is leveraged for the purpose of monitoring and recording states. Furthermore, the integration of Constrained Application Protocol (CoAP) with digital twin serves to reduce the frequency of packet exchange and retransmission, thereby enhancing the success rate to 97%. The test results include a comparative analysis of five different scenarios, which demonstrate that the proposed method yields a reduction in the performance of inactive twins to less than 1% when compared to the existing approach.

Shitharth Selvarajan, Hariprasath Manoharan

Chapter 8. Multi-Objective and Constrained Reinforcement Learning for IoT

Abstract

IoT networks of the future will be characterized by autonomous decision-making by individual devices. Decision-making is done with the purpose of optimizing certain objectives. A multitude of mathematically oriented algorithms exist for solving optimization problems. However, optimization in IoT networks is challenging due to a number of uncertainties, complex network topologies, and rapid changes in the environment. This makes the data-driven and machine learning (ML) approaches more suitable for effectively handling IoT environments’ dynamic and intricate nature. However, supervised and unsupervised ML approaches depend on training data, which is not always available before training. In recent years, reinforcement learning (RL) has attracted considerable attention for solving optimization problems in IoT. This is because RL has the distinguishing feature of learning with experience while interacting with the environment without training data. A central challenge in decision-making in IoT networks is that most optimization problems consist of co-optimizing multiple conflicting objectives. With the development of multi-objective RL (MORL) approaches over the last two decades, there is great potential for utilizing them for future IoT networks. Most recently developed MORL approaches have not been applied in the IoT domain. In this chapter, we will discuss the need for efficient multi-objective optimization in IoT, the fundamentals of using RL for decision-making in IoT, an overview of existing MORL approaches, and, finally, the future scope and challenges associated with utilizing MORL for IoT.

Shubham Vaishnav, Sindri Magnússon

Chapter 9. Intelligence Inference on IoT Devices

Abstract

With the rapid advancement of artificial intelligence (AI), the proliferation of deep neural networks (DNNs) has ushered in a transformative era, revolutionizing modern lifestyles and enhancing production efficiency. However, the substantial computational and data requirements generated by Internet of Things (IoT) devices present a significant bottleneck, rendering traditional cloud-based computing models inadequate for real-time processing tasks. In response to these challenges, developers have increasingly turned to cloud offloading as a solution, despite the high infrastructure costs and heavy reliance on network conditions associated with this approach. Meanwhile, the emergence of SoCs has enabled on-device execution, particularly on high-tier platforms capable of effectively handling SOTA DNNs. This chapter offers a comprehensive review of intelligent inference approaches, with a specific emphasis on reducing inference time and minimizing transmitted bandwidth between IoT devices and the cloud. The review encompasses various aspects, including the background of inference, hardware architectures supporting inference, a diverse range of intelligent applications, inference libraries tailored for IoT devices, and different types of inference techniques for applications. Additionally, this work addresses the current challenges in intelligent inference, discusses future development trends, and provides future research directions.

Qiyang Zhang, Ying Li, Dingge Zhang, Ilir Murturi, Victor Casamayor Pujol, Schahram Dustdar, Shangguang Wang

Chapter 10. Applications of Deep Learning Models in Diverse Streams of IoT

Abstract

Internet of Things (IoT) has gained enormous popularity in recent years. From obvious home automations to sophisticated medical procedures, IoT has gained considerable attention and applicability. But there are certain challenges also pertaining to apt use of IoT applications. The challenges range from generation of huge amount of data by sensors to security and privacy threats to IoT models. Malwares, energy consumption, and decision-making in healthcare or agriculture are few of the challenging aspects of IoT applications. The need of the time is to make IoT intelligent. Deep learning undoubtedly paves the way to put intelligence into IoT devices. Application of deep learning techniques helps IoT frameworks to handle difficult challenges more easily. For instance, deep learning models are very suitable to handle huge amount of data to find valuable inferences. Malware detection or optimisation of energy consumption in IoT applications finds right bid for deep learning models. In this chapter, we have gathered and compiled the applications of deep learning models in various fields of IoT. This chapter presents an in-depth study of these techniques in order to explore new horizons of applications of deep learning models in different areas of IoT.

Atul Srivastava, Haider Daniel Ali Rizvi, Surbhi Bhatia Khan, Aditya Srivastava, B. Sundaravadivazhagan

Chapter 11. Quantum Key Distribution in Internet of Things

Abstract

This chapter aims to explore the integration of Quantum Key Distribution (QKD) with IoT systems. It delves into the potential benefits, challenges, and practical considerations of incorporating QKD into IoT networks. The goal is to understand how QKD can enhance the security of IoT communication and enable a new generation of secure and trusted IoT applications.

Somya Rathee

Chapter 12. Quantum Internet of Things for Smart Healthcare

Abstract

The quantum Internet of things is an emerging paradigm that combines quantum technologies with traditional Internet of things systems to address the growing demands of smart healthcare. The integration of quantum computing, quantum communication, and quantum sensing into healthcare IoT devices promises unparalleled advantages, including enhanced data security, faster data processing, and improved accuracy in diagnostics and treatment. The advantages of quantum Internet of things in healthcare, such as enhanced data security, faster diagnostics, personalized medicine, and drug discovery, offer significant benefits to both patients and healthcare providers. Secure quantum communication protocols will ensure the privacy and integrity of sensitive medical data, paving the way for seamless telemedicine and remote patient monitoring. This chapter aims to provide a comprehensive overview of quantum Internet of things current state in the context of smart healthcare, including its applications, advantages, challenges, and future prospects. It explores the key components of quantum Internet of things for healthcare applications, examines the challenges and opportunities, and presents future directions for research and implementation in this transformative field.

Kartick Sutradhar, Ranjitha Venkatesh, Priyanka Venkatesh

Chapter 13. Enhancing Security in Intelligent Transport Systems: A Blockchain-Based Approach for IoT Data Management

Abstract

In intelligent transport systems (ITS), an interconnected Internet of things (IoT) device operates autonomously, collecting and exchanging data without human intervention. While the ITS offers numerous benefits, it also introduces various cybersecurity risks. These risks include the potential for undetected malicious IoT sensors, vulnerabilities in authorization and authentication protocols, and insecure data management practices. Therefore, it is crucial to develop new solutions that enhance the security of IoT sensors to address these challenges effectively. One specific challenge is securely managing and storing IoT data in intelligent transport systems. Current centralized systems are susceptible to tampering, necessitating more reliable data traceability records. This chapter proposes a blockchain-based architecture for securing and managing IoT data in ITS, offering advantages like immutability, decentralization, and enhanced security. Our results show that the proposed architecture performs well and can efficiently store vast amounts of IoT data. Additionally, it allows seamless addition and removal of new IoT sensors without retrofitting. Moreover, it is versatile and applicable to any intelligent transport system application. In conclusion, this architecture provides a robust and secure solution, significantly enhancing the overall security of IoT devices in this domain.

Chinmaya Kumar Dehury, Iwada Eja

Backmatter

Titel: Learning Techniques for the Internet of Things
herausgegeben von: Praveen Kumar Donta
Abhishek Hazra
Lauri Lovén
Verlag: Springer Nature Switzerland
Electronic ISBN: 978-3-031-50514-0
Print ISBN: 978-3-031-50513-3
DOI: https://doi.org/10.1007/978-3-031-50514-0