Hands-On Deep Learning for IoT

▶Book Description
Artificial Intelligence is growing quickly, which is driven by advancements in neural networks(NN) and deep learning (DL). With an increase in investments in smart cities, smart healthcare, and industrial Internet of Things (IoT), commercialization of IoT will soon be at peak in which massive amounts of data generated by IoT devices need to be processed at scale.

Hands-On Deep Learning for IoT will provide deeper insights into IoT data, which will start by introducing how DL fits into the context of making IoT applications smarter. It then covers how to build deep architectures using TensorFlow, Keras, and Chainer for IoT.

You'll learn how to train convolutional neural networks(CNN) to develop applications for image-based road faults detection and smart garbage separation, followed by implementing voice-initiated smart light control and home access mechanisms powered by recurrent neural networks(RNN).

You'll master IoT applications for indoor localization, predictive maintenance, and locating equipment in a large hospital using autoencoders, DeepFi, and LSTM networks. Furthermore, you'll learn IoT application development for healthcare with IoT security enhanced.

By the end of this book, you will have sufficient knowledge need to use deep learning efficiently to power your IoT-based applications for smarter decision making.

▶What You Will Learn
- Get acquainted with different neural network architectures and their suitability in IoT
- Understand how deep learning can improve the predictive power in your IoT solutions
- Capture and process streaming data for predictive maintenance
- Select optimal frameworks for image recognition and indoor localization
- Analyze voice data for speech recognition in IoT applications
- Develop deep learning-based IoT solutions for healthcare
- Enhance security in your IoT solutions
- Visualize analyzed data to uncover insights and perform accurate predictions

▶Key Features
- Understand how deep learning facilitates fast and accurate analytics in IoT
- Build intelligent voice and speech recognition apps in TensorFlow and Chainer
- Analyze IoT data for making automated decisions and efficient predictions

▶Who This Book Is For
If you're an IoT developer, data scientist, or deep learning enthusiast who wants to apply deep learning techniques to build smart IoT applications, this book is for you. Familiarity with machine learning, a basic understanding of the IoT concepts, and some experience in Python programming will help you get the most out of this book.

▶What this book covers
- Chapter 1, End-to-End Life Cycle of IoT, discusses the end-to-end life cycle of IoT and its related concepts and components, as well as the key characteristics and issues of IoT data that demands the use of DL in IoT. Furthermore, it also covers the importance of analytics in the IoT and the motivation to use DL in data analytics.

- Chapter 2, Deep Learning Architectures for IoT, provides the basic concepts of DL architectures and platforms, which will be used in all subsequent chapters. We will start with a brief introduction to machine learning (ML) and move to DL, which is a branch of ML based on a set of algorithms that attempt to model high-level abstractions in data. We will briefly discuss some of the most well-known and widely used neural network architectures. Finally, various features of DL frameworks and libraries will be discussed, which will be used for developing DL applications on IoT-enabled devices.

- Chapter 3, Image Recognition in IoT, covers hands-on image data processing application development in the IoT. First, it briefly describes different IoT applications and their image detection-based decision making. This chapter also briefly discusses two IoT applications and their image detection-based implementation in a real-world scenario. In the second part of the chapter, we shall present a hands-on image detection implementation of the applications using a DL algorithm.

- Chapter 4, Audio/Speech/Voice Recognition in IoT, briefly describes different IoT applications and their speech/voice recognition-based decision making. In addition, it will briefly discuss two IoT applications and their speech/voice recognition-based implementations in a real-world scenario. In the second part of the chapter, we shall present a hands-on speech/voice detection implementation of the applications using DL algorithms.

- Chapter 5, Indoor localization in IoT, discusses how the DL techniques can be used for indoor localization in IoT applications in general with the aid of a hands-on example. It will discuss how to collect data from those devices and technologies, such as analyzing Wi-Fi fingerprinting data through the use of DL models to predict the location of the device or users in indoor environments. We will also discuss some deployment settings of indoor localization services in IoT environments.

- Chapter 6, Physiological and Psychological State Detection in IoT, presents DL-based human physiological and psychological state detection techniques for IoT applications in general. The first part of this chapter will briefly describe different IoT applications and their decision making abilities based on the detection of physiological and psychological states. In addition, it will briefly discuss two IoT applications and their physiological and psychological state detection-based implementations in a real-world scenario. In the second part of the chapter, we shall present a hands-on physiological and psychological state detection implementation of the applications using DL algorithms.

- Chapter 7, IoT Security, presents DL-based networks and devices' behavioral data analysis, along with security incident detection techniques for IoT applications in general. The first part of this chapter will briefly describe different IoT security attacks and their potential detection techniques, including DL/ML-based ones. In addition, it will briefly discuss two IoT use cases where security attacks (such as a DoS attack and DDoS) can be detected intelligently and automatically through DL-based anomaly detection. In the second part of the chapter, we shall present a hands-on example of DL-based security incident detection implementations.

- Chapter 8, Predictive Maintenance for IoT, describes how to develop a DL solution for predictive maintenance for IoT using the Turbofan Engine Degradation Simulation dataset. The idea behind predictive maintenance is to determine whether failure patterns of various types can be predicted. We will also discuss how to collect data from IoT-enabled devices for the purpose of predictive maintenance.

- Chapter 9, Deep Learning in Healthcare IoT, presents DL-based IoT solutions for healthcare in general. The first part of this chapter will present an overview of different applications of IoT in healthcare, followed by a brief discussion of two use cases where healthcare services can be improved and/or automated through well-supported IoT solutions. In the second part of the chapter, we shall present hands-on experience of the DL-based healthcare incident and/or diseases detection part of the two use cases.

- Chapter 10, What's Next – Wrapping Up and Future Directions, presents a summary of the earlier chapters, and then discusses the main challenges, together with examples, faced by existing DL techniques in their development and implementation for resource-constrained and embedded IoT environments. Finally, we summarize a number of existing solutions and point out some potential solution directions that can fill the existing gaps for DL-based IoT analytics.