15.2.1 Identify the Benefits of Using Edge Computing from Different Perspectives

When studying edge computing and its applications, one cannot do so without understanding the opportunities and challenges that it presents to the issue of having centralized ICT solutions. A significant number of studies have highlighted motivations, benefits and challenges of edge computing [5–9]. Figure 15.1 demonstrates the motivations in edge computing which can be rendered as potential benefits of edge computing [5]:

Edge computing is critical for the Internet of Things (IoT) because of the added benefits that it can present to an IoT appliance. The IoT is defined as a system characterized by many interconnected objects that contain unique addressing scheme and sensors that are dynamically linked over a distributed network [6]. Edge devices that are dependent on the IoT include Amazon Alexa, Google Home, Google Assistant and all connected home appliances and devices. On the one hand, competitors might be racing for a low-latency processed data delivery system. On the other hand, consumers are expecting quick responses to their queries through their devices. Cloud-dependent services are challenged when it comes to low-latency computing. At this point, edge computing brings light to the problem of low-latency computing. Computing at the edge nodes is expected to be helpful to decrease dependency on the centralized cloud computing servers. There is a significant advantage of using edge nodes nearest to the edge devices to overcome network delays [5].

Edge devices can capture data from different sources. Data that are transmitted over the Internet to cloud computing servers for analyzing and analyzed outcome are transmitted back to the edge devices. In this type of condition, edge devices may experience high latency. Hardware and middleware limitations in edge devices may lead to failure while performing analytics which is complex in nature [7]. However, edge computing could be a beneficial solution to overcome resource limitations of edge devices, if a small proportion of data can be analyzed at edge nodes. Edge nodes can be used as unoccupied computational resources for analyzing or filtering data [5].

The benefits of edge computing also encompass a potential reduction of energy consumption at data centers and reduce the processing load of data centers. Energy consumption by data centers is projected to become three times greater than what is consumed nowadays [8], and energy-efficient approaches are profoundly required to diminish the consumption of energy [9]. Edge nodes within the edge networks can take a possible number of loads offloaded by cloud-based data centers according to capacity. According to Varghese et al. [5], consolidation of power management policy can be helpful to mitigate the challenge of energy consumption when edge nodes next to data sources can be used to handle some of the analytical tasks. Additionally, this may help reduce overloading problems of data centers.

It is obvious that edge devices are increasing rapidly. From mobile phones to wearable devices, a significant amount of data is being collected, which is increasing over time exponentially. Edge computing can be beneficial against increasing amount of data generated every day by a growing number of edge devices. An enormous amount of data which different devices generally generate is required to be handled in this world of interconnected networks. This situation triggers a need for data center expansion and raises a concern of increased energy consumption. Additionally, there is an increasing volume of traffic being forwarded to cloud-based servers which is another concern. Use of edge nodes could be potentially beneficial to reduce loads of cloud-based servers. Probable use of edge nodes to support some parts of computations of devices or data centers can be an advantage to cope with the surge of data and network traffic [5].

15.2.2 Identify the Applications of Edge Computing in Different Fields

Researchers show different types of applications of edge computing. According to the following studies, edge computing would open the door of possibilities:

1. Artificial Intelligence and Face Recognition Systems

   There are concerns of high-latency issues when data from face recognition systems are transmitted to the cloud for analyses purposes. A study was conducted by Zeng, Li et al. [10] where a face recognition system was developed to study the capabilities of Artificial Intelligence (AI) when applied to edge devices. A complete system had been built to prove the effectiveness of the proposed framework using the cloud server, rk3288 development platform and webcam [11].

2. Gaming

   Numerous computer or mobile phone users are playing video games, especially users with smart devices. Google has kept its promise to support gamers by launching Google Stadia in 2019, and now it is available in many countries. Although it has been reviewed as a technical and conceptual disaster, this is just a start. Google is rendering video games in their cloud servers and streaming videos to any devices at the user end. There is a major challenge for Google to overcome high-latency and bandwidth issues. Zhang, Chen et al. [12], proposed a framework to improve cloud gaming with the help of edge resources. Rendering part of edge games which requires demanding computation resources can be offloaded to edge devices helping to alleviate latency and consumption of bandwidth [12].

3. Healthcare

   Healthcare industry generates a large volume of data, especially emergency departments. Healthcare systems can be combined with edge computing technologies to increase their efficiency. A smart healthcare framework based on edge computing is proposed by Oueida et al. (2018). According to them, hospital emergencies deal with real-time systems which have complicated and dynamic structure, and the proposed framework offers the application of different key performance indicators optimized for a particular user group [13] (Figure 15.2).

   According to the framework stated above, all databases and healthcare-related software are deployed on the cloud where data is closer to the workflow of Smart HealthCare. Smart devices used in healthcare are edge devices. When assigned tasks are complete, a status notification containing the availability of resources in the pool is sent to the cloud servers with the help of edge nodes [13].

4. Oil and Gas Industry

   Oil and gas industry is expanding due to high demand for petroleum. Petroleum extraction plants use different types of sensors for extraction operations. According to Hussain, Salehi et al. [14], sensors for measuring pipeline pressure, gas leakage, air pollution and many more are used in extraction sites, and data is sent to the cloud servers for processing and high-latency satellite communications are used. All the latency-prone tasks or processes can be potentially handled with the help of edge computing [14].

5. Manufacturing

   In smart manufacturing plants, different types of tools, robots, machines, computers and devices relate to sensors to measure pressure, humidity and heat. In smart manufacturing, data is processed on the cloud servers, and factors like high latency, bandwidth issue and Internet unavailability affect the manufacturing process. Edge computing can be beneficial to reduce the issues stated above (Figure 15.3).

   According to Qi and Tao [15], edge computing responds to the issues mentioned above. The capacity of data computation, networking and data storing are extended by edge computing. Therefore, edge computing can be beneficial to manufacturing processes.

15.3.1 Issues in Security and Privacy

Cloud data center tasks are offloaded through edge computing with its own storage and evaluation in the edge network. Hence, performing the computation in edge network-level mainly raises the concern of providing data security and privacy. Information privacy and security have turned into fundamental prerequisites to ensure that clients in e-commerce businesses, financial and operational data are not compromised. This is a pressing issue with edge computing, and it provides an opportunity where Blockchain can play an integral part in addressing this limitation in edge computing. Additionally, when addressing privacy and security related to edge computing strategies need to be put in place for each tier of the edge processing frameworks. In the subsection that follows, embryonic privacy and security challenges dependent on the different tier engineering of edge computing will be covered. Figure 15.4 demonstrates the issues related to privacy and security in terms of edge computing.

1. Security for the core Infrastructure

   The core infrastructure of cloud storage and administration supports the edge paradigm. Different cloud service providers may be included inside of the core infrastructure to support the different edge layers. The core architecture will be supported by suppliers or any other provider (e.g. mobile network operators). Hence, privacy leakage, information altering, denial of service (DoS) attacks and administration control types of severe difficulties would arise due to semi-trusted or unsecured infrastructure.

   Edge devices will transmit different information in the main cloud storage through the edge data centers. The data that is transmitted would contain personal and confidential user information generated from different endpoint edge devices. Intruders might be able to tamper with the data or bypass the data processing while the edge data center is communicating with the core architecture. This highlights the issues of privacy leakages and information altering that exists with edge computing.

   Additionally, a malignant virtual machine can attempt to drain the assets where it is executing, and for this reason, it can also hamper computational processing, network stability and storage assets. According to Zhang, Chen et al. [16], there is a severe threat as most edge data servers will not have the information which can be accessed by other cloud frameworks. Inside of the distributed and decentralized architecture, a hijacked or jammed core infrastructure network can provide or exchange false information due to insufficient service privileges [16]. This prevents unauthorized access to sensitive data. Figure 15.5 enlisted the categorization of privacy and security challenges in an edge computing architecture.

2. Security for Edge Servers

   Different virtualized services and management of services are executed through edge servers. Like the multi-cloud server scenario, the edge server performs its activities in a specific geographical location. Considering the case, here, both external and internal parties can access or alter the private data which are processed by the server. Accessing sensitive data or false data injection attacks may happen due to a lack of privacy within the system. Consequently, the server activities can be manipulated by intruders if it has breakable access privileges. The full access can make the edge data server a rogue data center as in extreme cases; it will control the whole infrastructure. Another threat includes a physical attack of the actual edge data center. Even though the edge servers require a specific data center storage location, to be identified, however, these, servers may be physically damaged and/or stolen [17]. It is worth noting that attacks are restricted to certain locations because of the decentralized nature of edge computing.

3. Security for Edge network

   According to the paradigm of edge computing, it will reduce the possibilities of severe attacks in the network, such as the distributed denial of service (DDoS) and DoS attacks. Malicious activities may hamper the network through attacks like eavesdropping and traffic injection. There are also high chances of a man-in-the-middle attack where a hacker can hijack the streaming information of the network. Malicious activities may be launched through the rogue gateway, which can have a similar effect to that of a man-in-the-middle on the whole infrastructure [16]. In such cases, the network infrastructure is compromised, and hackers can easily gain access to sensitive information that is transmitted over the network.

4. Security for the Edge Devices

   With IoT, devices are interconnected with each other over multiple communication networks such as wireless networks, Internet and mobile core networks. In edge computing architecture, there are numerous IoT-related gadgets at each compositional layer. These devices are inclined to actively interact with each other at the edge ecosystem, and these could be constrained by a central malignant framework. The attack brought about by malware comes in numerous shapes and types. Malware can capture data and sensitive information from smart devices such as debit and credit card user information. Further, malware can obstruct the elements of physical savvy hardware which can have a severe impact [18].

5. Privacy-Preserving

   The multi-layered architecture of the edge computing framework creates the possibilities of cybersecurity and privacy issues. Such issues can be adequately addressed using state-of-the-art systems for access control that would use sophisticated algorithms and cryptography at the edge of the network to restrict unauthorized user access. The traditional access control system will not be compatible with these more dynamic and agile environments of the edge paradigm. The aforementioned different malicious activities can be controlled through a sophisticated access control system.

   Edge computing paradigm includes services, for instance, data storage, virtual machine, and the total infrastructure encapsulates edge devices, edge data center, and core infrastructure. Multiple functional roles are executed by end-users, service providers and infrastructure providers in the edge architecture [16]. In the complex situation of the edge environment, it is not only about assigning every component in one trust area but also about the need to let the elements commonly confirm each other among various trust areas. Thinking about the high versatility of the edge gadgets, the handover confirmation innovation is likewise a significant examination point in the validation convention. Secure lightweight multi-factor authentication protocols are a more realistic solution for edge framework to preserve privacy [19, 20].

   Additionally, an enormous security transformation is occurring because IoT devices are normally not providing built-in security mechanisms. This gives intruders easy access to the system. Because getting remote access to the edge layer much of the time breaks confidentiality. One of the significant issues is data integrity for the security of the edge network. As the client information is transmitted to the servers of the edge network, the information trustworthiness could be undermined executing this procedure. In the data transmission procedure, the owner should check the integrity and accessibility of the information to ensure that there are no undetected alterations of information by any unapproved user. The protected information search is the preferred test which implies the client needs to tackle the issue of the keyword search over the encrypted information records [20–22].

15.3.2 Issues in Decentralized Architecture

In decentralized conditions, where any client can turn into a supplier and assets might be heterogeneous, contrasting equipment and network execution is essential. The optimal test for decentralized cloud arrangements is confirmation of the calculation, which is expected to keep away from malevolent activities by suppliers and customers. End-user information can be overseen by decentralized cloud systems. It is essential that solutions to resolve information security are thoroughly considered, and that necessary actions are taken to mitigate such hazards. In this case, probably the verification process is a vital challenge for decentralized edge computing with the cloud server. The targeted approach to resolve this issue would be to restrict the different malicious activities performed by customers and service providers [23, 24].

In a distributed architecture the network has heavy traffic congestion as endpoint devices transmitting data to the cloud server. Consider the performance of a system; it might be affected by the remoteness of the communication between the endpoint devices with the main server. Moreover, the volume of data being transferred between different applications such as live audio streaming, video streaming or e-commerce would be dependent on the location of the main cloud server. With the use of edge servers, edge computing can decrease the transmission latency by sieving and managing the data. This ensures that less bandwidth is used, and it improves the energy efficiency of the system. Hence, when applied to smart applications, data processing would be faster and more efficient through edge computing in a decentralized framework [25].

15.4.1 Blockchain: Beyond Cryptocurrency

Blockchain is a rapidly emerging technology that involves a distributed chain of blocks or transactions that provides publicly available immutable records for the participants (e.g. human, device) [26]. In the Blockchain, the existing records cannot be modified or erased as each block is connected using a cryptographic hash of another block, along with a timestamp and transaction data. New blocks are appended only at the end. Therefore, Blockchain is also known as the immutable data storehouse [22]. Though the history of the application of Blockchain starts with cryptocurrency (e.g. bitcoin), nowadays it is not limited in cryptocurrency but applicable across different industry and sectors within the economy and society as a whole. Due to the benefits of Blockchain technology, it has become a widely used technology in many sectors, especially in the smart grid and IoT [27]. Figure 15.6 illustrates the key application areas of Blockchain.

15.4.2 Advantages of Blockchain

There are several reasons behind the wide adaptation of Blockchain technology in different fields. Blockchain technology has the ability to ensure integrity, transparency, high security, simplicity, faster processing power and above all, decentralization.

1. Decentralization

   The main advantage of Blockchain technology is its decentralized architecture. Instead of storing valuable data at a central location, Blockchain technology stores data across its peer-to-peer network, and it reduces the risk of storing data at a central location by eliminating numerous threats. In this decentralized environment, it is more difficult for intruders or hackers to infiltrate any vulnerabilities in the network.

2. Process Integrity and Transparency

   Process integrity and transparency are two of the advantages of Blockchain technology. The technology is designed in such a way so that each record or transaction is added to the Blockchain and the data of these entries are available to the participants who were involved in the transaction. Once these transactions or blocks are added to the chains, then it cannot be modified, which results in a high level of process integrity in Blockchain compared with other approaches [28–30]. Additionally, transparency is attained by copying the transactions to all the computers connected to the Blockchain network. Hence, the participant can access all the public transactions, which cannot be modified [31].

3. Security

   Ensuring better security is another great advantage of Blockchain. When a user enters the Blockchain network, the user is given a unique identity that is linked to his/her account. This identity ensures that the transaction is performed by the owner of that account. In addition, the use of a cryptographic hash enhances the security of this technology further where the cryptographic hash of the previous block is held by the newly created block that leads to creating the chain. This hash is created automatically that consists of a type, ID number of the block, time of creation, previous hash’s value, miner’s level and Merkle root [30]. This enables higher security levels to be maintained in Blockchain. Furthermore, the combination of the Blockchain hashing process and cryptography leads to immutability that can be defined as the ability of a Blockchain ledger to remain unchanged.

4. Simplification

   Blockchain offers a simplified ecosystem of transaction over the existing multiple ledger system. In the existing system, multiple ledgers result in disorder and difficulties to the parties involved with the system. By offering a single public ledge to store all the transactions, Blockchain is a simplification of the ecosystem [30].

5. Faster Processing

   Another advantage of Blockchain is increased speed in processing. In a conventional approach, a transaction would take up to 3 days in processing and authorization. However, using the Blockchain technology, it takes a few minutes as Blockchain helps to reduce time through secured digital signature [30].

6. Privacy

   The risk of a data breach has increased with the wide adaptation of technology in recent years. Such data breaches pose a threat to violate the privacy of the customers/users as these data may contain delicate private data such as the name of a person, identification number, banking details and their residential or work address. Traditionally, a provider such as a bank or a hospital would be responsible for creating the data and storing it in their corporate database. However, this creates several security and privacy concerns that require immediate attention [32]. Blockchain technology has the ability to address these problems to ensure high security and privacy due to its decentralized structure which ensures that sensitive information can be better protected [32, 33].

15.4.3 How Blockchain Will Complement Edge Computing

Based on earlier discussions, the main challenges of edge computing are distributed control, ensuring security and preserving privacy. This highlights a clear gap in current edge computing approaches and the need for a more secure form of a transaction at the edge of the network. This is where Blockchain technology can be advantages because of its decentralized nature and its ability to ensure security and privacy preservation. Delivering cloud resources and services at the edge of the network is the main motivation behind edge computing. However, there are many limitations that need careful considerations to ensure that the costs do not outweigh the benefits [4]. Yang and et al. [4] observed that integrating Blockchain with edge computing would allow distributed control over numerous edge nodes. Moreover, Blockchain technology can maintain the accuracy, consistency and data validation on a large number of nodes throughout their life cycle [4, 34]. It also ensures privacy by storing data locally or among multiple parties as small fragments which reduces the risk of disclosing large amounts of information due to the coordination of information via its ledge system. This challenge can be overcome with the help of Blockchain technology as it offers complete privacy through manageable keys to each user for accessing and controlling data and coordination without revealing any metadata to the peers on the network. It is highly likely to maintain privacy by integrating Blockchain technology with edge computing. Figure 15.7 portrays the advantages of Blockchain technology and disadvantages of edge computing.

15.4.4 How Blockchain Can be Integrated with Edge Computing

Integrating Blockchain with edge computing has several benefits. Several architectures or frameworks are proposed in the existing studies. However, before integrating Blockchain with edge computing, there are some requirements that need to be fulfilled. In this section, these requirements are discussed followed by the summary of the concepts, frameworks and challenges covered in this section.