Preface

Fog computing is a decentralized computing structure that connects data, devices, and the cloud. It is an extension of cloud computing and is also known as edge fog networking, or fog networking, or fogging. Fog computing is an essential concept in IoT (Internet of Things), as it reduces the burden of processing in cloud computing. It brings intelligence and processing closer to where the data is created and transmitted to other sources.

Fog computing has many benefits, such as reduced latency in processing data, better response time that helps the user’s experience, and security and privacy compliance that assures protecting the vital data in the cloud. It also reduces the cost of bandwidth, because the processing is achieved in the cloud, which reduces network bandwidth usage and increases efficiency as user devices share data in the local processing infrastructure rather than the cloud service.

Fog computing has various applications across industries, such as agriculture and farming, healthcare industry, smart cities, education, and entertainment. For example, in agriculture industry, a very prominent example is the SWAMP project, which stands for Smart Water Management Platform. With fog computing’s help, SWAMP develops a precision-based smart irrigation system concept used in agriculture, minimizing water wastage.

This book is divided in to three parts. The first part studies fog computing and machine learning, covering fog computing architecture, application perspective, computational offloading in mobile cloud computing, intelligent cloud-IoT system, machine learning fundamentals, and data visualisation. The second part focuses on applications and analytics, spanning various applications of fog computing, such as in healthcare, Industry 4.0, cancer cell detection systems, smart farming, and precision farming. This part also covers analytics in fog computing using big data and patient monitoring system, and the emergence of fog computing with reference to applications and potentialities in traditional and digital educational systems. Last but not least, the third part covers security aspects in fog computing through blockchain and IoT, and fine-grained access though attribute-based encryption for fog computing.

Part I: Study of Fog Computing and Machine Learning

Chapter 1 starts with an overview and introduction to fog computing characteristics followed by a brief description of fog computing’s application in intelligent Cloud-IoT systems. Then a detailed fog computing architecture is given, with descriptions of basic modules in this architecture. Following that, a relative comparison of cloud computing and fog computing is drawn by examining the applications of fog computing. The chapter ends with a summary of challenges faced by, and the future scope of fog computing.

Chapter 2 discusses how computation offloading is a critical technology in the rapidly developing field of Mobile Cloud Computing (MCC). MCC can improve application speeds, reduce latency, and extend battery life. Among other things, the effect of computation offloading is influenced greatly by a variety of parameters. After a short introduction and summary of related work, this chapter leads with different computation offloading techniques. The chapter ends with related studies and comparisons of these offloading techniques, along with the future scope of research in mobile offloading.

Chapter 3 explores the optimized and green uses of fog computing in Industry 4.0. After an introduction to Industry 4.0 and fog computing, this chapter focuses on how IoT integration with fog computing will have wide application in Industry 4.0. This chapter also provides two fog computing architectures, hierarchical and layered. This chapter ends with descriptions of various applications for fog computing, plus data analysis with related figures and tables.

Chapter 4 elaborates upon machine learning and its integration in agriculture. This chapter starts by demonstrating the importance of integrating machine learning solutions into agriculture. Then it proceeds to demonstrate fog computing as a backbone for collecting and filtering agricultural data. Afterward, a proposed model is given as methodology, followed by a discussion of the results and various modelling algorithms, such as Decision Trees, Random Forest, XGBoost, CatBoost and LightGBM.

Chapter 5 discuss how smart buildings utilize Internet of Things (IoT) devices, sensors, software, and internet connectivity to monitor various aspects of the building, analyse data, and extract insights to enhance the building’s environment and operations, i.e., the role of intelligent IoT applications in fog computing. Following that is a detailed discussion of fog and cloud computing, and fog computing with IoT, Edge, AI, and other network intelligence objectives. The chapter concludes with case studies.

Chapter 6 gives a SaaS-based data visualization platform in Covid-19 perspective. The chapter starts by defining the pandemic in terms of data visualization and Power BI, followed by a summary of data collection and wrangling. A proposal for design and implementation are given, followed by dashboard development. The chapter ends by addressing advantages, impact, and future scope.

Chapter 7 is a complete study of machine learning algorithms for medical data analysis, particularly COVID–19. After an introduction, this chapter explains pre-processing of medical data for machine learning, followed by supervised learning and Support Vector Machine (SVM). Next, Naive Bayes Algorithm and KNN are covered. Thereafter, the chapter covers deep learning algorithms with illustrations of unsupervised learning, and applications of ML algorithms in medical data analysis. All are illustrated with medical data analysis. The chapter ends by addressing future research scope and challenges.

Part II: Applications and Analytics

Chapter 8 illustrates an application of fog computing in healthcare industry. The chapter starts by presenting a layered architecture of fog computing in detail, followed by research methodologies. Then this chapter gives an application taxonomy of fog computing-based healthcare system with disease diagnosis, monitoring, and notification. The chapter ends by considering the challenges and research opportunities of a fog computing-based healthcare system.

Chapter 9 gives an IoT-driven predictive maintenance approach to industry 4.0 with a Fiber Bragg Grating (FBG) sensor application. After a detailed introduction, different ML algorithms are reviewed, such as LDA, DT, kNN, SVM, and RF. Thereafter, research gaps are highlighted with emergent research directions. The chapter ends by covering the broad concept of FBG sensor applications in Industry 4.0.

Chapter 10 proposes a fog computing enabled cancer cell detection system using Convolution Neural Network in Internet of Medical Things (IoMT). After an introduction to fog computing in IoMT, the chapter discusses the relationship between IoMT and Deep Neural Network. The chapter then proposes a model of fog computing enabled CNN for Medical Imaging, followed by an algorithmic approach to proposed models. The chapter concludes with results and analysis.

Chapter 11 gives a detailed, application-based review of smart and precision farming. After the introduction, the chapter starts with methodologies used in precision agriculture, mainly map-based and sensor-based techniques and the contribution of IoT. Then a detailed study of IoT-enabled smart farming and precision farming is provided. Following that is an explanation of machine learning-based precision farming, illustrated with a case study of OR methods in farming systems. The chapter ends with conclusive remarks and a future scope.

Chapter 12 provides analytics of big data obtained from fog computing, with an introduction, literature review, and a summary of motivation. The chapter details fog computing architecture, followed by a thorough discussion on big data, then examines the big data analytics, and ends with concluding remarks.

Chapter 13 reviews the application of a IoT-based patient monitoring system in real time. After an introduction, the components used for this model are discussed, such as Node MCU, Heart Rate/Pulse Sensors, and Temperature Sensors (LM35), followed by a discussion of IoT platforms, such as ThingSpeak. A method is proposed, followed by instructions for experimental setup and results. The chapter concludes with the outcomes of this model and an explanation of how to use IoT-based patient monitoring system effectively in real time.

Chapter 14 is a scientific review of fog computing and its emergence, with reference to applications and potentialities in traditional and digital educational systems. The chapter starts with a background study, followed by a summary of the methods used for this research and the basics and advantages of fog computing. Following that is a report on the increase of fog computing applications and its impact on education and IoT security.

Part III: Security in Fog Computing

Chapter 15 illustrates the use of blockchain security for fog computing. This chapter gives thorough examples and outlines detailed security issues that present in fog computing environments. Following that is a detailed explanation of blockchain architecture, further discussion of the topic, and summary conclusions.

Chapter 16 covers an extension of blockchain security in IoT with fog computing. This chapter starts with a detailed discussion about...