Preface

This book intends to increase knowledge regarding emerging techniques of Artificial Intelligence (AI)-based Internet of Things (IoT), its diverse applications, and probable security threats for IoT devices. The Internet of Things (IoT) is a critical component of Industry 4.0 and is sometimes used interchangeably. This is an emerging technology that enables legitimate users to access and monitor the sensors installed in various units in the industry. The smart industry is being developed with the help of IoT. It is used in many applications like smart cities, smart parking, digital healthcare, smart agriculture, smart disaster management, and many more. The data produced by the IoT devices is semi-organized or non-structured, requiring cloud on-demand storage to store that data. The IoT aims to build automated systems so industry and society can proliferate. The number of IoT devices used to implement applications is growing at an exponential rate.

Moreover, device-to-device connectivity makes IoT devices more vulnerable. Thus, secure and authenticated communication is in demand for IoT devices. Furthermore, different AI-based techniques, including fuzzy logic, machine learning, and neural network-based approaches, are proposed for early predicting any severity in the IoT paradigm. Adding an AI-based multi-stage decision-making mechanism also helps optimize the assistive resource distribution where IoT integrates with cloud and fog computing. The resource-constrained IoT devices demand lightweight security schemes to authenticate communicating devices and protect the privacy of industrial data. In the manufacturing industry, the sensor nodes of the IoT network are deployed in a hostile environment. Thus, it is not feasible for the industry to keep track of those devices, and it will be easier for unauthorized users to access the smart devices of the IoT. Thus, this book's editors and contributing authors tried to cover some of the AI-based IoT techniques with an elaborated description of their applications and also propose schemes to prevent vulnerabilities in different security threats.

Chapter 1 overviews blockchain technology for smart IoT, its architecture, security challenges, and applications. This chapter first highlights the introduction of smart IoT with security and privacy concerns for these systems, followed by a detailed systematic analysis of blockchain architecture, its characteristics, consensus algorithms, various platforms, and application areas in smart IoT to analyze the concept and working of blockchain technology. Various benefits and challenges in integrating blockchain with IoT have been analyzed and illustrated.

Chapter 2 describes the role of blockchain in today’s financial growth using IoT. Through this analytical survey, the authors have tried to analyze the decentralization that improves consensus success and how these descriptions of blockchain redesign the financial domain and the landscape of rivalry among several crypto-currencies. It has various applications with highly increasing demands for the best solution in a distributed environment using IoT in the present market scenario.

Chapter 3 provides a survey of forest fire surveillance strategies and challenges using the WSN paradigm. The forest is the most beautiful treasure in nature. It always meets the basic needs of the earth's inhabitants. Today, the forests are depleting quickly. The primary cause behind this is forest fires or wildfires. An uncontrollable fire occurs naturally or due to human interruption or any other disturbance caused by nature that may or may not be suppressed by artificial control. Several existing approaches, like wireless sensor networks, machine learning, and remote sensing, are used to identify wildfires. Some researchers are using UAVs to identify forest fires. In most cases, the researchers focus only on prediction using some environmental parameters sensed by the sensors or satellite images. This chapter highlights the various challenges in predicting forest fires in the WSN paradigm.

Chapter 4 proposes an IoT-based intelligent emergency alert system using neural computing and machine learning. Fuels, gases, etc., have prominent uses in our daily lives, households, and industries. However, they often cause severe accidents from gas leakage and fire incidents. The authors have designed a simple system using low-cost devices that sends an SMS via a GSM module in case of gas leakage or fire using IoT, neural computing, and ML. The objective of the proposed system is to enhance safety and security and protect properties.

Chapter 5 offers an IoT-enabled framework for secure and transparent digital answer script evaluation using blockchain. The proposed method proved to be beneficial due to its decentralized nature. In the proposed framework, all the steps of evaluation processes are traced and recorded through the blockchain to improve the system's security, transparency, and trustworthiness. This makes it simpler to figure out how a candidate obtained the score that he or she did, providing credibility to the certificate obtained.

Chapter 6 proposes an intelligent farm management system using an IoT-based Agrobot to match pH, temperature, humidity, and soil moisture levels to the levels required for growing crops. Productivity is increased without negatively impacting the soil. This reduced the use of chemical fertilizers by 30% to 75%. This proposed system may help to improve economic and environmentally sustainable crop production.

Chapter 7 provides a novel, unbiased trust establishment mechanism in a cross-domain, cloud-based IoT environment. The authors advocate a trust-dependent authorization model to be implemented in a cloud-dependent IoT environment that also functions as a two-way or dual-mode trust model, considering the requirements of the service giver and taker. In the suggested paradigm, trustworthiness is assessed on both the user and supplier sides. A transaction is only permitted if both trust values exceed a predefined or set threshold.

Chapter 8 uses graphical representation and deep learning to classify IoT malware network traffic. This revolutionary technique of IoT malware traffic analysis uses deep learning and graphical demonstration to detect and categorize new malware more quickly. This will allow us to handle the difficulty that has been presented (zero-day malware). Due to the utilization of deep learning technology, the suggested method for detecting malicious network traffic operates at the package level, significantly reducing the time required for detection and producing encouraging outcomes.

Chapter 9 describes the optimum utilization of modern-day technology for health monitoring with wearable devices using IoT. The Immediate Health Monitoring System uses IoT, enabling it to monitor the patient's temperature along with the oxygen level instantly; this system transmits the same information to the doctor or medical service provider at a distant location. It also enables one to look at the patient's current condition. If the values of the parameters change from the traditional values, then an alert message is given to the medical service provider or the doctor concerned with the patient. This instant health monitoring plan based on IoT helps doctors effortlessly collect real-time numbers at a location far from the patient's location.

Chapter 10 proposes an IoT-enabled automated model for detecting COVID-19 spread using deep learning. The proposed model diagnoses COVID-19 cases quickly and maintains low hardware costs. The proposed model has been tested on 280 COVID-positive and 290 normal patient images taken from two benchmark datasets and has tried to find out the spreading tendency of the virus by getting some actual data from IoT sensor devices in the affected zone. The proposed model achieves an accuracy of 97%. Moreover, cross-validation is applied in this model to avoid over-fitting. This chapter concludes that the proposed model provides insight into the CNN-LSTM+Capsule network used for COVID-19 detection, allowing for richer feature mapping from radiographic images and effectively distinguishing COVID cases from normal ones.

Chapter 11 describes a model for the automated identification of cloud-IoT-based sensitive data in a dataset. Before sharing data with a third party, it is anonymized and de-identified. The detection of data points that have the potential to expose sensitive information can be a tedious task, especially if done manually. Automating the task helps make identification much more manageable when dealing with many small and large data sets. The current solution has been envisaged to help identify potential leakage of sensitive information using an easy-to-implement framework and a solution for detecting potential quasi-identifiers in data.

Chapter 12 proposes an IoT and blockchain-based smart architecture for secure supply chain management. This chapter discusses the architecture of different entities involved in the agriculture supply chain and their roles and interactions with other entities. In this case, smart contracts are used between the farmer and the customer, the farmer and delivery personnel, and the farmer and the quality assurance authorities involved in the agriculture supply chain. It also discusses their roles and interactions with other entities.. The authors built the system using the Remix IDE and Solidity for programming smart contracts.

Chapter 13 suggests the optimization of the application-aware QoS Routing Algorithm (AQRA) and MINA for SDN-based IoT networks that ensure numerous QoS specifications for high-priority IoT applications. The Particle Swarm Optimization (PSO) technique has been applied to get the optimized solution.

Chapter 14 describes IoT-based data security in smart farming systems. The security system is created using a few sensors, like temperature and soil moisture sensors, and the data is encrypted. The sensors transform the plain text into unintelligible cipher text, which is then uploaded to the cloud on the transmitter side. The AES128 key and hash code are used on the transmitter side, which can be used to decode data on the reception side.

The editors would like to express their gratitude to all the reviewers for their outstanding contributions to this book. We sincerely hope that readers enjoy reading these chapters, and we anticipate that they will aid in advancing IoT techniques and implementation research. This book will be a huge success regarding idea exchange, leading to future research collaborations in emerging AI-based IoT techniques and applications.