Abstract
Internet of things (IoT) has revolutionized digital transformation and is present in every sector including transportation, energy, retail, healthcare, agriculture, etc. While stepping into the new digital transformation, these sectors must contemplate the risks involved. The new wave of cyberattacks against IoT is posing a severe impediment in adopting this leading-edge technology. Artificial intelligence (AI) is playing a key role in preventing and mitigating some of the effects of these cyberattacks. This chapter discusses different types of threats and attacks against IoT devices and how AI is enabling the detection and prevention of these cyberattacks. It also presents some challenges faced by AI-enabled detection and prevention and provides some solutions and recommendations to these challenges. The authors believe that this chapter provides a favorable basis for the readers who intend to know more about AI-enabled technologies to detect and prevent cyberattacks against IoT and the motivation to advance the current research in this area.
TopIntroduction
Internet of Things (IoT) is another technical revolution whose time has also most definitely come (Atzori et al., 2010). Internet of Things (IoT) is the interconnection of heterogeneous smart devices through the Internet with diverse application areas such as smart home, car, cities, healthcare, wearables, retail, grid, agriculture, and the industry as shown in Figure 1. Gartner, a global research and advisory firm, forecasts that the IoT market will grow to 5.8 billion endpoints in the year 2020, a 21% increase from the previous year. Most of the enterprises have already started embracing IoT as a way of expediting the digital transformation initiatives. The essence of IoT will give all companies a new norm of digitizing business models. The IoT will enable a smooth integration of business processes through digitization. However, one of the main challenges plaguing a successful deployment of IoT is Cybersecurity. In IoT, not all of the Things that form an IoT network have adequately tested for cyberattacks. As a result, the entire system is threatened.
Cybercrime is increasing, and cybersecurity is ultimately evolving. This evolution is calling for innovation in the field of cybersecurity defense. This phenomenon is becoming more vital due to the boost on the Internet and diversified devices that constitute the IoT. The contemporary world of digital transformation is bringing challenges to the cybersecurity of the IoT environment. A cyberthreat is a threat that involves a computer or a computer network. Vulnerabilities in computers bring about cyberthreats. Vulnerabilities are due to weakness in the design, implementation, operation, or control of computers and computer networks. Because of the exploitable vulnerabilities, cyber hackers attack computing systems of individuals or cooperation. As a result, we need to defend these computing systems against cyberattacks.
Figure 1. Internet of Things: the big picture
Until recent decades, traditional defense mechanisms are deployed to protect the computing facilities. However, history showed that traditional defense mechanisms are not adequate anymore as hackers become more sophisticated and plentiful. Thus, the Artificial Intelligence (AI) might be the only way to keep pace. Artificial intelligence (AI) is currently being implemented into information systems across all industries. A vast amount of data is being produced regularly by IoT devices. The data produced by IoT devices is far more than any traditional method that can process or make use of in a productive way. As a result, the IoT environment is progressively challenging cybersecurity professionals to secure without hindering IoT functions. Fortunately, AI is a solution. The IoT obligates AI. AI is a key cybersecurity weapon in the IoT era. In this work, we introduce how AI enables cybersecurity for IoT.
Around the 1990s, because of low speed, the use of the Internet was very limited where the Internet connectivity was only diffused into the enterprise and consumer market. Around the 2000s, the Internet brought up many applications in our lives. By 2010, Internet connectivity propagated into the enterprise, consumer, and industrial products to provide access to all kinds of data and information from the sensor and to actuator devices that form the concept of IoT. The evolution of the IoT from “Internet of the Content” to “Internet of Things” is depicted in Figure 2. Nevertheless, today, these IoT devices are still only “Things” on the Internet, as they require human interaction and monitoring through applications. The future is promising. IoTs responding to how we want everything to act and operating ubiquitously behind the scenes. For sure, life is going to be amazing.
AI is a field of computer science addressing to automate computing problems that require human intelligence (Aishath et al., 2019). The AI tries to construct artificially learning and acting agents by researching living creatures existing in nature. In other sophisticated words, it aims to create agents that have high cognitive functions like sensing, learning, thinking, associating, reasoning, problem-solving, communicating, inferring, and decision-making capabilities. If general AI is to be accomplished, it might also lead to superintelligence. Since both AI and IoT lie on a large amount of data, it is one of the best fits for the Cybersecurity needs of IoT.
Key Terms in this Chapter
Information Gain: Information gain is the reduction in entropy or surprise by transforming a dataset and is calculated by comparing the entropy of the dataset before and after a transformation.
Edge Computing: It is a distributed computing paradigm that brings computation and data storage closer to the location where it is needed, to improve response times and save bandwidth.
Deep Neural Networks: A deep neural network is an artificial neural network with multiple layers between the input and output layers.
Artificial Intelligence: AI is a field of computer science addressing to automate computing problems that require human intelligence.
Internet of Things: Internet of things (IoT) is the interconnection of heterogeneous smart devices through the Internet with diverse application areas such as smart home, car, cities, healthcare, wearables, retail, grid, agriculture, and industry.
Fog Computing: Fog computing or fog networking, also known as fogging (edge computing), is an architecture that uses edge devices (e.g., IoT devices, etc.) to carry out a substantial amount of computation, storage, and communication locally and routed over the internet backbone.
Artificial Neural Networks: Artificial neural networks, usually simply called neural networks, are computing systems vaguely inspired by the biological neural networks that constitute animal brains.
Convolutional Neural Network: Convolutional neural network is a class of deep neural networks that mostly applied to analyzing visual imagery. They are generally used in image recognition, classification, face recognition, etc.
Cybersecurity: Cybersecurity is the protection of computer systems and networks from the cyberattacks so that their hardware, software, or data are not disrupted or misdirected of the services that they provide.
Cybercrime: It is a crime that involves a computer and a network. The computer may have been used in the commission of a crime, or it may be the target. Cybercrime may threaten a person, company or a nation's security and financial health.
Blockchain: It is a growing list of blocks that are linked using cryptography. Each block contains a cryptographic hash of the previous block, a timestamp, and transaction data. A blockchain is resistant to modification of its data. It is primarily used in cryptocurrencies, most notably bitcoin.