Introduction

In 2012, Cisco introduced the concept of Fog Computing (FC) to improve network infrastructure to match the demands of the large amounts of data being transmitted to the cloud for processing [1]. That is, FC was introduced to help and overcome the problems faced by cloud computing (the use of the Internet to supply on-demand computer services such as storage, apps, and processing capabilities) such as the connectivity between the cloud and the Internet of Things (IoT) devices, the latency-sensitive applications, location awareness of the IoT applications, and complexity of the distribution environment [30]. Seven categories of the BC with FC integration purposes were identified; security, privacy, access control, trust management, data management, scalability, and performance. The paper also presents a roadmap of prospective research areas, problems, and possibilities for which more studies are required to guide the researchers. This was done by addressing the limitations of reviewed papers and identifying some open issues in infrastructure, platform, and technical limitations of BC architecture that distress processes in specific realms. It’s important to note that this analysis is by no means comprehensive since BC technology continues to advance at a breakneck speed. The rest of this paper is organized as follows. Research bacground presents an overview of BC. Blockchain overview discusses the research methodology. Blockchain with fog computing integration overview discusses the descriptive findings. Research methodology discusses BC with FC integration purposes. Locating studies discusses the future challenges and open questions about BC with FC integration. Study selection and evaluation concludes with options for further research.

Research background

Blockchain overview

BC can be defined as a distributed append-only public ledger technology that was originally proposed for cryptocurrencies (e.g., Bitcoin) [24]. In 2008, the concept of BC was proposed by [31]. Transactions occur among different parties without the supervision of a central authority. The valid transactions, using the consensus mechanism, are then recorded in the ledger (chronologically blocks that form a BC) and copied to all parties. A consensus algorithm is used to construct blocks and add them to the ledger which sometimes represents a computational issue. Three considerations are required for BC construction; immutable ledger, transparent and public ledger, and anonymity of the BC users [18].

The majority of the background body was built using bitcoin BC, which is the first and most widely used BC platform among a wide range of applications. Another reason for discussing Bitcoin BC in greater depth rather than other platforms such as Ethereum (a decentralized open-source BC with smart contract capability that is most recognized for its native cryptocurrency, ETH, ether, or just Ethereum) is the extensive literature accessible on the platform [32]. Bitcoin BC, for example, uses SHA-256 hashing and elliptic curve cryptography to provide robust cryptographic evidence for data integrity and authentication [20]. The elliptic curve cryptography is a key-based encryption system that employs pairs of private and public keys to encrypt and decrypt data [14]. The BC, usually, includes a list of all transactions and a hash to the prior block, which enables a cross-border distributed trust environment. While trusted parties or centralized authorities may misbehave and can be compromised, disrupted, or hacked, transactions in the public ledger of BC are validated by a majority consensus of miner nodes involved in the validation process [33]. In PoW-based BCs, for example, the validation occurs by calculating a hash with leading zeros to meet the difficulty target [20]. After validating by a consensus, the transaction data are saved in a ledger that not be erased or changed (data are immutable) [34].

Figure 1 describes a typical structure of the Bitcoin BC which consists of a sequence of blocks connected through the hash value. The BC includes the block header and the block body includes the transactions list. Various fields are included in the block header such as the block size, a timestamp, the number of transactions, and the version number. The hash value of the current block is represented by the Merkle root field. Hashing using the Merkle tree is often used in Peer-to-Peer (P2P) and distributed arrangements as it provides effective data proof. The nonce field is included as a Proof-of-Work (PoW) algorithm (the original consensus algorithm in BC (e.g., Bitcoin and Ethereum), which is used to confirm transactions and produce new blocks in the chain), and it is used to generate the trial counter value that generates the hash with leading zeros [32]. The number of leading zeros is specified by the difficulty target (i.e., used to preserve the block time of nearly 17.5 s for Ethereum and 10 min for Bitcoin [20]). The difficulty target can be modified to increase the number of zeros if the computation power of the hardware increased. The timestamp is used for tracking the modification on the BC. Different mechanisms are used for timestam** such as signing using the private key of a trustworthy server used in the traditional schemes [35]. Another technique can be used by deploying distributed timestam** which helps to avoid a single point of failure [35].

Fig. 1
figure 1

Bitcoin BC structure

Full size image

The method by which a BC network achieves consensus is referred to as a consensus mechanism or algorithm. Since there is no central authority, the public BC (i.e., decentralized) is constructed as a distributed mechanism, with distributed nodes agreeing on the validity of transactions using a consensus algorithm [34]. In other words, BC depends on distributed consensus to validate the transactions which guarantee the consistency and integrity of the transactions [36]. The different consensus mechanisms impact the BC system differently [37]. The best (idealistic) consensus mechanism promotes giving the same weight to all miners for the validation process and then deciding based on the majority. This ideal scenario may be applicable in a controlled (private) environment; however, in public contexts, this may increase the chance of Sybil attacks as users can share multiple identities [35]. In distributed architecture such as FC, only one random user will add every block which may lead to several attacks [38].

Bitcoin is the most well-known cryptocurrency. Later, in 2015, Ethereum BC was launched, which can execute smart contracts and store data [38]. The smart contracts are programs written and uploaded by parties to be executed in the BC which includes the terms of the contract. Soon later, other BC platforms were launched such as Stellar (a digital money protocol that’s distributed and open-source), Hyperledger (a worldwide business BC initiative that provides the structure, tools, and rules for creating open-source BCs and apps), Ripple (a BC-based digital payment system and mechanism with its cryptocurrency, XRP), Eris (an open-source software that enables anybody to create low-cost, safe, and portable apps utilizing smart contract and BC technology), and Tendermint (an algorithm for securely and consistently replicating applications over many devices) [20, 21, 32]. Depending on the data managed, the availability of that data, and the actions taken, different types of BC can be identified. It is worth mentioning here that some authors refer to public/permissionless and private/permissioned, interchangeably. This can be applicable in cryptocurrencies; however, in other applications that need to distinguish between authentication and authorization, it’s not applicable. Though, the naming is still in debate among authors. Note that Bitcoin, for instance, is used to track digital assets, while smart contracts used in Ethereum enable certain logic. Moreover, while some system like Ripple makes use of tokens, others like Hyperledger do not.

In general, BC can be categorized into three major types; public (e.g., Bitcoin, NXT, CounterParty, RootStock, and Zcash platforms), private (e.g., Monax, Hyperledger Fabric, Ripple, Multichain, and Corda platforms), and consortium (e.g., Ethereum, Monax, and Multichain platforms) [23, 24, 39]. 1) On the Internet, everyone can see the public BC ledgers, and anybody may validate and contribute a transaction to the BC. 2) Only selected people inside the company may add and validate transactions in Private BC, but anybody with access to the Internet can normally read them. 3) Consortium BC allows only a group of organizations (e.g., financial institutes) to add and validate a transaction, however, the ledger might be available or limited to certain parties. Applications such as auditing within an organization and data management require consortium BC, in general, as public BC is not suitable for user privacy and commercial benefit protection [23].

Accordingly, BC offers the following benefits over other technologies [34, 36]. 1) Resilient - no single point of failure and using smart contracts, which means BC helps in transferring, securely. It is a network of nodes, all nodes work collaboratively to maintain the transaction, records are augmented to a ledger of a previous transaction, and PoW should be validated by other nodes included in the chain. 2) Decentralized and trustless-P2P system, which cuts the need for any kind of agent for security by cryptography. The distributed database is duplicated into every node, which includes timestamps, transaction lists, and information with links to the previous blocks in the chain. The distributed ledger should be transparent, immutable, publicly accessible, and updated after each transaction. 3) Scalable and high speed and capacity technology. The computing capacities of the network scale up when a new peer joins the chain. 4) Secure and transparent because every transaction is visible to every miner on the chain.

While a lot of research has been conducted on BC technology, the state-of-the-art of BC with FC integration purposes has received insufficient attention [40]. The main impetus for this work was the lack of a clear and complete analysis of existing BC with FC integration purposes state-of-the-art in the literature. BC can avoid many attacks even without centralized control or data storage [23]. The Ethereum-transaction-based state-machine provides special features like security, transactional privacy, integrity, authorization, auditability, data immutability, fault tolerance, and transparency [24]. Accordingly, many applications use this technology nowadays rather than cryptocurrencies such as smart transportation, identity management, industry, agriculture, energy grids, supply chain management, and FC [22].

Blockchain with fog computing integration overview

FC is a highly dispersed computing structure with a set of assets made up of one or more pervasively linked embedded systems (which include IoT devices) supported by cloud computing, to cooperatively offer storage, computation, storage, connectivity, and other services to a sizable number of IoT devices nearby [3]. FC is a cloud expansion that is more closely connected to IoT devices. FC serves as a bridge between edge devices (e.g., sensors, and actuators) and the cloud [14]. A fog node could be any device having processing power, storage capabilities, and network connection, including routers, security cameras, switches, and control devices. Distribution, flexibility, proximity to IoT devices, low latency, real-time transactions and analysis, and heterogeneity are typical characteristics of FC [41]. All of these qualities made FC a very alluring remedy for cloud computing problems, particularly excessive latency and centralized authority [42].

Many studies have been conducted recently that discussed the value of BC in an FC environment such that devices like personal computers, mobile units, and Vehicular Ad-hoc Network (VANET) can be equipped with BC. The role of BC in FC can be broadly seen from two angles; data processing and communication [43]. That is, the role of BC will be very important in maintaining security and privacy on the fog nodes when data is stored or processes in the fog node and when data is transferred between fog nodes, between fog nodes and the cloud, and between fog nodes and the IoT devices. The fog node will play the operator role (i.e., manage) for IoT devices [14]. The decentralized and dispersed fog nodes, associated with the network, handle the communications included in BC. Each block in the BC is attached to the chain sequentially [34]. All nodes included in the BC environment are parts of the network which store a local copy of the transaction data permanently. All the parties involved jointly authenticate the transaction to meet a consensus decision, before a miner node (e.g., Ethereum Virtual Machines - nodes that can provide trustworthy execution cryptographically tamper-proof and administration to these contracts or programs) add the validated transaction into a timestamped block [20]. And then broadcasts it into the network. This data is periodically updated among all nodes for consistency purposes. This enables many nodes, that do not trust each other, to achieve authentication decisions based on the old transactions. In the BC environment, a public ledger preserves the validated transactions in a P2P network. In general, two keys are used: 1) a private key which is used to sign the BC transaction before broadcasting to other peers and 2) a public key that represents the unique address [18].

In order to obtain BC incentives, nodes compete in PoW to perform cryptographic formulas and verify transactions. On the other hand, Proof-of-Stake (PoS) employs random selection validators to guarantee the transaction’s dependability and pays them with cryptocurrency [44]. The most popular cryptocurrency, Bitcoin, employs PoW. The second-largest cryptocurrency, Ethereum, began off with PoW but is now switching to PoS. High levels of reliability and security are stated for PoW [45]. The intricacy of the mathematical calculations required to attain verification makes manipulating the system all but useless. But it’s slow and expensive to run, and it consumes a lot of energy. PoS eliminates the need for difficult calculations. Instead of figuring out a numerical riddle, the miner in PoS-based BC employs a digital signature as evidence. Instead of receiving a newly formed asset, the miner who verifies the block is compensated with a transaction fee [46]. PoS consensus maintains the incentive mechanism and effectively assures node equity since it has a low relative burden on computational resources and high throughput. By examining the quantity and duration of tokens it has, PoS calculates the likelihood of acquiring accounting privileges [47]. Similar to the stock dividend system, people who possess comparatively greater shares might get higher dividends. Therefore, it is more energy-efficient than PoW and provides higher sustainability [48]. The nodes with stakes are meant to be trustworthy and refrain from manipulating transactions, but if they do, their stake might be taken away. Participating in the PoS is simpler for investors than the PoW since it doesn’t need technical skills or computer-aided design. PoS outperforms PoW in terms of speed as well. For instance, Ethereum can handle up to 100,000 transactions per second using PoS, but it can only handle 30 transactions per second with PoW [48]. In the case of PoS, however, there is a possibility that a node will not have enough assets, in which case, if it were to be chosen as a miner, it would be viewed as malicious since it would have no assets to be debited [47].

Research methodology

To identify and synthesize the purposes of integrating BC in FC, we adopted a Systematic Literature Review (SLR) approach based on the guidelines provided by [49, 50]. SLR aims to identify, select, and synthesize the available literature to answer the research question [30]. A systematic literature review protocol is essential to guide the review process [30] that provides a framework to understand the impact of BC on FC security and privacy challenges. We have developed a review protocol to validate the classification process of this paper. Distinct stages have been applied: (1) locating studies, (2) screening studies, (3) study selection and evaluation, and (4) study inclusion.

Locating studies

The following seven well-known electronic databases were used in this review. These databases are expected to provide enough literature coverage for this paper.

In the first stage, all possible combinations of BC, FC, and edge computing were searched using the Boolean “AND” and “OR” operators. The edge computing term was included in the search terms because many authors refer to FC as edge computing. The selected studies come from different IoT applications of FC such as vehicular, smart cities, and health applications. The selected papers include peer-reviewed articles published in journals, book sections, or conference proceedings. Figure 2 shows the stages of the review process and the number of papers identified at each stage. In this review, we included any study that discussed BC as a technique used in fog or edge computing. Therefore, studies were excluded if their focus was not on fog or edge computing or if they did not discuss using BC. This review included studies up to April 2022; qualitative, quantitative, mixed measurement studies, overview studies, and review studies. The search excluded studies that discuss prefaces, poster sessions, editorial discussion, news, article summaries, or reader’s letters. Only papers written in English were included.

Fig. 2
figure 2

Study selection process

Full size image

Study selection and evaluation

The authors individually evaluated all of the literature using the established criteria, as discussed in Section 3.1. All authors sat together, at the end of each stage, and discussed the included and excluded studies. In this review, we followed the citation procedure discussed in Alzoubi et al. [50]. We used EndNote as a citation manager tool to store the selected studies. Moreover, we used the backward snowball sampling technique and searched the reference lists of the selected studies, in the first stage, to get new studies. The number of hits resulting from the first stage was 517. After excluding the non-English written studies, the number dropped to 508. Moreover, the number dropped to 501 after excluding the duplicated papers.

The 501 papers were imported to EndNote (to keep track of the references) and Excel sheet (to maintain the abstracts and titles). In this stage, the titles of the selected studies were reviewed. The papers that were not about BC with FC integration were excluded. However, some titles failed to be identified, and thus included in the next review stage. In this stage, 374 studies were identified as relevant to the scope of this study. Moreover, after reviewing the abstracts of the selected papers, the number dropped down to 187 papers. The abstracts that were not considering any application of BC with FC integration were excluded (e.g., architectural and/or technological features of BC). Some abstracts were misleading so the papers, in this case, were included in the next stage. If the abstract was not available, the study was left for stage 4. At stage 4, all potential studies were gone under the full-text review. In this stage, 6 papers were excluded as they did not report the BC with FC integration, leaving 181 papers for the final inclusion stage.

Data extraction and synthesis

All articles that matched the requirements for inclusion were entered into MAXQDA11, a qualitative analysis program, and the data was evaluated for emergent themes. The thematic analysis for selected papers was conducted independently by the authors. In the end, the seven categories were compared among all authors. The consensus rate was around 78%. All authors agreed on all articles included for thematic analysis (N = 181), one set of categories, and sub-categories. The selected studies [38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218] are discussed in the following Sections. First, a descriptive analysis is provided for the selected studies. Next, the taxonomy of the BC with FC integration’s positive impact on security and privacy issues of FC is presented. Finally, the future directions of this SLR are discussed.

Descriptive analysis

The study looks at 181 academic articles that were published between 2016 and April-2022. The descriptive analysis serves several aims including fascinating insights into current research patterns in BC technology. It also serves as a guide for future studies. Moreover, it aids in visualizing the interdisciplinary research techniques that have been established in the scientific literature thus far. Table 1 summarizes the studies that were chosen based on the published database. IEEE was the single largest publication outlet, with 104 studies out of 181 (74 journal articles and 30 conference proceedings), followed by Elsevier Science Direct with 21 research. As the smallest number of studies, just two were retrieved from SAGE. The “IEEE Access” journal, which published 19 papers, was found to be the single most popular publication channel. Table 1 further reveals that the bulk of the papers chosen (132 out of 181) were peer-reviewed journal articles, followed by 43 conference proceedings, and only six-book sections.

Table 1 Publication channel
Full size table

Figure 3 shows a year-by-year examination of the selected publications. It’s worth mentioning that the number of publications was low in 2016 (1 study) and 2017 (7 studies), but increased in 2018 to 26 studies. However, in 2019, the total number of studies hit a high of 49. The number drops to 44 studies in 2020, 43 studies in 2021, and 11 studies by April 2022. This trend reflects the fact that BC technology is new and develo**, as well as the increasing scholarly interest in it. Even though BC technology was initially established using Bitcoin as a basic underlying innovation and Bitcoin has accounted for the majority of investigated platforms over the last seven years, many of the papers published in 2020, 2021, and 2022 focused on the latest or modern BC platforms, such as Etherium, with a particular focus on smart contracts. Figure 3 also shows that the vast bulk of the literature was published in peer-reviewed journals, with only a few book sections.

Fig. 3
figure 3

Publication year-distribution

Full size image

BC originally started with Bitcoin (BC 1.0), then BC 2.0 which was built on smart contracts, and later evolved into coordinative applications (BC 3.0) [35]. The majority of BC with FC integration applications discussed in selected studies were IoT applications (83 studies), transportation (31 studies), eHealth (16 studies), industrial IoT (9 studies), monitoring applications (6 studies), energy (8 studies), mobile devices (4 studies), supply chain management (4 studies), drones’ network (3 studies), video streaming (2 studies), financial (2 studies), global collaboration (2 studies). Other applications were also revealed including FC-PoW approach [45], higher education applications [207], FC-resource brokerage platform [134], FC-authentication scheme [218], agricultural supply chain [156], multi-party contract signing [217], video streaming [99], consensus for edge-centric IoT [57], intelligent and safe task offloading in vehicles [89], and FC-rogue nodes approach [135]. Other review papers focused on BC with FC integration in general ([63, 85, 109]) and FC security ([45, 200]). The other studies which included literature review are not counted in these applications.

Figure 4 shows the domain-purpose distribution of the 181 research across time. BC with FC integration has been divided into seven domains based on the results of the research. Security takes up the most research items (38 out of 181), followed by performance (34 studies), trust management (31 studies), privacy (25 studies), access control (21 studies), data management (16 studies), and lastly scalability (11 studies). Figure 4 demonstrates that, even though BC with FC integration is still in the early phases, its goals have expanded beyond security and privacy to include trust management, data management, performance, and scalability concerns. Furthermore, a significant number of publications addressing the subject of trust management were published in 2019 (12 studies). Moreover, the focus among the selected studies has been more on enhancing FC-IoT-cloud architecture using BC technology. It’s worth noting that several authors highlighted the role of BC in FC as a supplement to security and privacy concerns. In other words, they assumed that, by default, BC enhances the security and privacy of FC, then can achieve other purposes such as trust management, performance, or scalability. As a result, when classifying the results in this paper, we focused on the primary goal of each study.

Fig. 4
figure 4

Domain year-distribution

Full size image

Twenty two papers were literature review includes one paper published in 2016 ([125]), one paper in 2017 ([57]), two papers in 2018 ([63, 185]), seven papers in 2019 ([27, 109, 164, 194, 207, 210, 228]), three papers in 2020 ([85, 86, 195]), five papers in 2021 ([60, 191, 192, 200, 211]), and three papers up to April 2022 [153, 154, 214]. Some of these papers focused on certain purposes such as resource management [154], while others were general literature reviews without focusing on certain purposes such as [214]. The latter, though, were not included in the classification of Fig. 4.

More information about the survey studies found in our systematic evaluation is provided in Table 2. While the bulk of these surveys concentrated solely on a single area of BC with FC integration, such as health or transportation, or a single purpose, such as security and privacy, this article offered a thorough analysis of all purposes and from all areas of literature that were accessible. Furthermore, unlike this article, none of the identified survey studies have systematically investigated BC with FC integration.

Table 2 Survey studies focus
Full size table

Blockchain-fog computing purposes

This paper focuses on BC with FC integration purposes. We suggest a purpose-oriented categorization in this paper. Our approach, on the other hand, varies from comparable studies (E.g., [154, 192, 211, 214]) as it does so by utilizing a rigorous statistical methodology based on the literature, making it more relevant to present BC advances and illustrating future BC trends with high fidelity. As a result, we present a thorough and comprehensive classification of BC-based goals, which is visually depicted in Fig. 5, taking into consideration the current and future variety of BC solutions. Based on an examination of the existing literature, we provide a thorough taxonomy of the BC-enabled purposes that are currently accessible in the following subsections. The purpose categories identified in this paper were, however, classified (coded) using the prior literature review publications as a starting point. Most research reviews, for example, identified security and privacy as the major purposes of BC with FC integration. We begin classifying with these purposes and then add the evolved categories like trust, performance, access control, and scalability. The coding was done according to the definitions given to describe each purpose category and its subcategories.

Fig. 5
figure 5

BC with FC integration purposes

Full size image

Security

Data can be harmed by a variety of security risks. BC may be able to shield you against these dangers to a large extent. Availability, confidentiality, and integrity are the most important security purposes [192]. We found several studies that indicated security support and fraud detection in addition to these three purposes. These purposes are discussed in the following sub-sections.

Security support

Many studies have reported that BC can enhance and support the security of FC, in general without focusing on a specific particular of security. Several new solutions were proposed to enhance security in the BC with FC integration environment. To provide an efficient and secure communication framework, Alam [176] emphasized the confluence of BC, FC, and IoT technology advancements. Similarly, Alam [177] presented a framework for delivering middleware on the Internet of smart devices network. The suggested framework is particularly well suited to applications in which data is sent regularly on the Internet of smart devices environment. Ashik et al. [139] created a FC-cloud architecture based on BC that may be utilized in smart homes. By leveraging the BC network, this design gives rise to a distinct fog architecture that provides greater security against known threats to safeguard our sensitive data. Dorri et al. [82] suggested a BC-based architecture to safeguard users’ privacy and strengthen the vehicular ecosystem’s security.

Huang et al. [81] proposed a distributed security approach using smart contracts and the lightning network; this suggested model is known as the lightning network and smart contracts. To improve the security of trade between charging piles and electric cars, the new suggested security model can be combined with existing scheduling software. Huang et al. [84] proposed a BC system to address the IIoT security problems. The authors also created a data authority management technique to control access to sensor data to safeguard sensitive data confidentially. Huang et al. [94] used BC technology to create a decentralized parked vehicle aided FC. Smart contract executions arrange and validate request posting, workload completion, task appraisal, and reward assignment automatically. This method provides strong security and efficiency guarantees, as demonstrated by a security study and comprehensive numerical findings [94].

Rahman et al. [104] demonstrated a safe therapeutic framework that allows patients to own and control their personal data without the assistance of a trustworthy third party, such as a therapy facility. With BC’s support, the framework can withstand unwanted access or a single point of failure. Although the BC only maintains the treatment metadata’s immutable hashes, the actual multimedia data, depending on the application’s needs, audios, videos, photographs, or other augmented reality therapeutic data is saved off-chain in a decentralized database. This functionality allows you to make use of metadata’s immutability while annotating or upgrading multimedia big data [104]. Shynu et al. [120] proposed a secure BC with FC integration healthcare service for illness forecast. When develo** projections, cardiovascular disorders are considered. The patient’s health data is initially gathered from fog Nodes and stored on a BC. When compared to existing neural network methods, the suggested approach achieved a prediction accuracy of over 81%.

Fraud detection

Fraud detection is the process of checking a document or other data system to see whether there has been any tampering with the data or other harmful activity [164]. The focus here is on how BC can protect FC from attacks. Jeong et al. [223] proposed creating a secure FC system using a reliable distributed BC. IP spoofing, Sybil attacks, and single point of failure may all be prevented with our suggestion. The digital signature utilized in the transaction creation process ensures authenticity and non-repudiation in this proposal. Because it is based on a BC, which is a distributed ledger, it can effectively restore or alternate a downed FC even when it is offline. Stanciu [149] presented a study based on the IEC 61499 standard that uses BC technology as a foundation for hierarchical and distributed control systems. Hyperledger Fabric was chosen as the BC solution, with function blocks being implemented on a supervisor level as smart contracts [149]. Liang et al. [222] suggest utilizing cross-BC-enabled FC to provide safe service detection for the Internet of Multimedia Things (IoMT). An extensible cross-BC design based on FC is provided initially to avoid tampering and espionage during the trust evolution process, in which separate parallel BCs may be coordinated to communicate hidden geographic data and app trusted proof. The smart contract in the BC-based Ethereum is meant to allow Turing complete computing [222].

Misra et al. [147] recommended using a private BC network to implement a Software Defined Networking (SDN) architecture in a fog-enabled IoT ecosystem to prevent such hostile attacks against controllers in real-time. If the miners discover incorrect flow rules, BC permits the SDN devices/fog nodes to revert to a previous flow rule while flagging the accused controller. The authors also recommended encrypting the data before placing it into the blocks, which would help protect the data from unauthorized users [147]. Moreover, Rathore et al. [161] BC technology was offered as part of an SDN-based decentralized security architecture. SDN is in charge of providing an optimal attack detection model by continuously monitoring and analyzing data. The single point of failure concern in the present design is mitigated by BC’s decentralized threat detection [161].

Gul et al. [168] proposed a business model for the healthcare industry that uses BC to link the FC and the cloud. Certain data in the healthcare industry can be analyzed for prediction, and companies can plan before disaster strikes. Many attacks are thwarted since there is no direct contact between the BC layer and the IoT layer. Because the company can predict the course of business and make decisions appropriately, this fusion makes business more productive. Kumar et al. [221] employed two Artificial Intelligence (AI) approaches, random forest and XGBoost, to offer the proposed security framework full autonomy in decision-making skills. An interplanetary file system is recommended for distributed storage and data load balancing. To identify DDoS assaults in smart contracts, the authors presented a distributed system based on FC. The suggested distributed framework’s findings demonstrate that it is extremely successful at identifying numerous assaults in the BIoT network, such as DDoS and other current attacks [221]. Kumar et al. [45] demonstrated how the integration of BC using the PoW consensus mechanism can enhance FC security.

Sharma et al. [73] presented a novel Distributed Mobility Management (DMM) solution based on BC technology for flattened FC. The suggested solution can deal with hierarchical security concerns while maintaining network layout. It uses three BCs to meet the needs of completely distributed security while also resolving the de-registration difficulties that plague previous DMM systems. Furthermore, the distributed BC approach aids in the prevention of DDoS, backward broadcasting attacks, session hijacking, and impersonation attacks. It also encourages the use of de-registration rules. Sivasangari et al. [181] presented a BC with FC integration design to identify security threats at the cloud layer, resulting in a reduction in IoT security attacks. The elliptic curve cryptography-based proxy encryption is used in the proposed design.

Confidentiality

Confidentiality refers to the assurances that the data may only be accessed by authorized users or nodes. Other nodes are unable to comprehend the private and secret information that each node possesses [192]. Farhadi et al. [224] explored how distributed BC ledger technology may be utilized to address Confidentiality, integrity, authenticity, non-repudiation, and availability challenges in FC architecture as decentralized computing support [224].

Gao et al. [48] provided a new framework called SGX in the IoT-cloud medical health (IoMT) using BC with FC integration to maintain a trusted environment and data confidentiality. To maintain the highest level of data protection, only a portion of the relevant diagnostic data can be given to the medical facilities in need. Curious data processing facilities, on the other hand, will potentially contribute to data leakage. FC and BC were combined to provide a new platform to address these issues. Mohapatra et al. [229] presented a secure data exchange system for IoT devices based on BC with FC integration. The authors proposed two software agents: a BC creation software agent deployed in FC, and a network of IoT device monitoring software agents. Block addition by an approved IoT device is done with an AES 128-based PoW while hashing in BC was done with SHA 256. To improve FC privacy, Wu et al. [123] integrated BC with FC and leveraged multi-party secure computing technique in smart contracts. Participants can only access the output value of their functions using this technique, which encrypts output and input. Simultaneously, the BC may verify and agree on the findings calculated by this technique across the whole network.

Integrity

Data integrity guarantees that the message’s content is not tampered with during transmission [211]. As a result, unlawful data production, deletion, or alteration is prohibited [192]. By allowing all network members to collectively own and validate data, which was previously handled by a centralized server, BC enhances transaction record integrity and dependability. The technology may minimize brokerage fees and construction expenses thanks to distributed data management, while also ensuring high levels of data integrity and security [61]. Kumar et al. [87] argue that the BC maintains data integrity, security, and trust in a decentralized manner. Accordingly, the authors have proposed the BlockEdge framework, which brings these two enabling technologies together to solve some of the existing IIoT networks’ most pressing challenges [87].

A BC-based crowdsensing framework was presented in Gu et al. [127] to deal with security risks, which helps validate the authentication of supplied sensor data and resists record tampering. Guo et al. [227] offered a lightweight encryption system with outsourced decryption. Encryption is the process of converting an original text or data into an alternate version known as ciphertext in order to ensure data confidentiality [32]. Although outsourced decryption reduces the data user’s computing overhead in an attribute-based encryption system, the ciphertext is uncontrollable, and the data owner cannot ensure the data’s accuracy. The proposal guarantees that ciphertext is verifiable, allowing the user to quickly verify for accuracy. Moreover, using BC, the authors enclosed the hash value of the public parameter, the original and modified ciphertext, as well as the transformed key into a block, allowing for tamper-resistance against both internal and external attackers [227].

Jang et al. [193] presented a novel BC with FC integration architecture for IIoT that prevents data falsification by changing existing centralized database methods to distributed types based on BC. They presented a technique to organically manage the IIoT ecosystem by splitting the proposed system structure into cloud, FC, and IoT devices. Users are transferred to the cloud to assure integrity, stability, and scalability. The authors recommended using a fog node to handle smart contracts and transaction verification to improve network latency (the required time for data to move from one location to another) and throughput. For 5G-enabled drone identification and flying mode detection, Gumaei et al. [58] proposed a system that integrates a Deep Recurrent Neural Network (DRNN) with BC. Raw RF signals from various drones in various flight modes are remotely detected and gathered on a cloud server to train a DRNN model, which is subsequently distributed to edge devices for identifying drones and their flight modes. The suggested framework uses BC to ensure data integrity and security [58]. Without a tamper-proof audit, centralized compute offloading poses a security risk. It was unable to protect against false reporting, free-riding, spoofing, and repudiation attacks. As a result, Huang et al. [94] used BC technology to create a decentralized parked vehicle aided FC. Smart contract executions arrange and validate request posting, workload completion, task appraisal, and reward assignment automatically. To reduce security threats, network operations in computation offloading become transparent, verifiable, and traceable [94].

Availability

Availability is a critical component of security services, assuring that the system and other apps continue to function in the event of a malfunction or hostile attack [192]. Muthanna et al. [204] proposed an IoT framework that uses a fog node layer managed by an SDN network to deliver high availability and reliability for delay susceptible applications. BC was used to guarantee that decentralization is done safely [204].

Current Agri-Food supply chain provenance and traceability applications are controlled by a centralized technology, which allows the opportunity for unresolved issues and key concerns, such as data integrity, manipulation, and single points of failure [137]. The transaction records are fault-tolerant, immutable, transparent, and fully traceable thanks to BCs [137]. Caro et al. [137] proposed AgriBlockIoT, a completely decentralized, BC-based traceability system for the Agri-Food supply chain that can seamlessly connect IoT devices that produce and consume digital data throughout the chain. They created and deployed such a use-case, establishing traceability using Ethereum and Hyperledger Sawtooth, two distinct BC implementations [137].

Insights and discussion

Due to the immutability of the BC, tampering with the data kept in the system is unlikely, and participants’ identities and data integrity may be assured. The data in the BC contains the whole transaction history, which is hashed to keep the ledger secure. As a consequence, BC can ensure that devices are connected (e.g., through smart contract-verified transactions). Fabricating data is almost impossible in the BC system due to the joint monitoring of linked fog nodes (i.e., the attacker will have to alter all of the data on the connected fog nodes, in order to fabricate the data). As a result, BC is protected by distributing data over a large number of linked fog nodes. Authors proposed several architectural designs to support security in FC environment ([81, 84, 104, 120, 139]): to protect against frauds ([73, 147, 149, 164, 168, 181, 221,222,223]), to enhance and achieve data confidentiality ([48, 123, 192, 229]), to enhance and achieve data integrity ([32, 58, 61, 87, 94, 127, 193, 227]), and to achieve data availability ([137, 192, 204]). The majority of the selected studies under this category reported that BC can help against fraud attacks in FC, followed by data integrity purpose, and the least purpose mentioned was to achieve data availability.

Privacy

Messages including identity, location, and other personal data are used by many apps and services. As a result, maintaining one’s privacy is critical. The rising demand for FC systems is creating a huge amount of sensitive data. This section discusses the privacy-related purposes including privacy support, identification privacy, data privacy, and location privacy.

Privacy support

Several studies have reported that BC can enhance the privacy of FC, in general, as follows. The use of Consortium BC in conjunction with the Transport Layer Security Protocol (TLSP) maintains security and privacy while reducing the requirement for a third party [143]. Pavithran et al. [169] proposed a privacy-preserving BC architecture for IoT. The proposed architecture is well-suited to event-driven IoT devices, and it makes use of the edge and cloudlet computing paradigms, as well as Hierarchical Identity Based Encryption (HIBE) for privacy protection, in which the ciphertext comprises only three group components, and decryption needs only two bilinear map calculations. Uddin et al. [162] suggested a decentralized eHealth architecture based on BC technology. To guarantee patient privacy while outsourcing duties, a patient agent program uses a lightweight BC consensus mechanism and a BC leveraged task-offloading algorithm [162].

Huang et al. [217] developed a fair three-party contract signing mechanism based on BC. To achieve fair trade, the suggested structure employs the verified encrypted signature and the BC. As a result, if a dishonest party aborts after obtaining the present product, it will be punished financially [217]. Gai et al. [93] developed a permissioned BC-edge architecture for smart grid networks to solve two fundamental smart grid concerns: security and privacy. To ensure the legitimacy of users, the authors employed covert channel authorization mechanisms and group signatures [93]. Smart contracts on the BC were used to create an ideal security-aware approach. The efficacy of the proposed technique has been validated for the proposed model [93]. Guan et al. [115] proposed a smart grid scheme for BC-based dual-side privacy-preserving multi-party computing. To assure the security of multi-party computing in edge nodes (e.g., summing), the scheme uses the data segmentation technique. To improve system security and eliminate reliance on trustworthy third parties, the consortium BC and smart contract were used [115].

A decentralized and privacy-preserving charging method for electric cars has been suggested by [88]. The BC system is installed on distributed FC nodes, allowing for a decentralized and secure storage environment. The privacy in the charging process may be maintained by integrating mutual authentication, smart contracts, and BC-based storage [88]. Nadeem et al. [175], in the CRVANETs ecosystem, presented an effective and secure BC scheme-based distributed cloud architecture. Instead of using traditional cloud architecture, on-demand sensing and minimal cost were used to protect the drivers’ privacy. The proposed architecture provides drivers with the necessary security for future autonomous driving [175].

Qu et al. [174] suggested a user-friendly FC architecture. According to the recommended design, clients enroll their devices in the fog portal which acts as an intermediary between the resources of each local network and the IoT service [174].

BC with FC integration can solve the problem of identifying, authenticating, and verifying healthcare IoT devices in a decentralized context [172]. Accordingly, Shukla et al. [172] proposed a new solution to the aforementioned dilemma, integrating FC and BC. This solution used the Advanced Signature-based Encryption (ASE) method (a type of digital signature that uses an enhanced certificate to verify the signer) for healthcare IoT device authentication [172]. Tang et al. [136] used a combination of BC and FC to verify each fog server’s identity and create a secure offloading system. A BC-based offloading mechanism was provided to reduce query time and offload security for potential fog servers. A BC-based technique, on the other hand, has inherent limits. All transactions should be recorded to a single copy BC database on each server. If a fog server can handle various queries at once, there will be a large amount of synchronization overhead as a result of this [136].

Wang and Jiang [218] proposed a 2-adic ring identity authentication system that inherits the 2-adic ring’s strong key distribution and great validation efficiency, and this algorithm includes trading node supervision and identity hiding functions by design. The consortium BC was used for this system [218]. Yang et al. [61] looked at how to manage identifiers effectively with BC technology in a named data networking context. By establishing a transaction using the identification’s content name, the suggested system does not reveal a specific user’s identifier. Using an identifier split management approach, the identifier may be safely kept and controlled [61]. Zhu and Badr [129] proposed a hybrid IoT architecture that combines FC with a trustless IoT environment to assure security. Users may easily manage smart devices by establishing tamper-proof digital identities and building a new class of authentication and authorization methods for the IoT by enabling this architecture with BC-based social networks. Fog nodes may also manage all IoT entities’ identities and relationships, as well as implement IoT security measures [129].

Data

To protect data privacy, we must ensure that only authorized nodes have access to the data. Data privacy is another important issue for FC [60]. Lautert et al. [146] proposed architecture for tracking data provenance in a distributed FC over a large region. Using software services that maintain the information consistent across all interested parties in the cloud, the architecture presented in this article allows quick and accurate data provenance for clients operating in the FC. The suggested architecture is based on the well-known W3C Prov provenance concept, which makes the framework easier to use. The authors created a client and web services application that allows users to store and exchange provenance information in a BC using open standards [146]. To protect IoT data, Liu et al. [188] presented a decentralized access control mechanism based on BC with FC integration. To encrypt IoT data before uploading to the cloud, this technique employs mixed linear and nonlinear spatiotemporal chaotic models, as well as the least significant bit. The evaluation showed that this mechanism can alleviate the problem of a single point of failure and ensures the privacy of IoT data.

In vehicular fog, there are still several issues with the secure and reliable transmission of sensory data. To address these concerns, Kong et al. [97] proposed a verifiable sensory data collecting and sharing method in vehicular FC using a permissioned BC. By integrating the homomorphic 2-disjunctive normal form cryptosystem with an identity-based signcryption method, the proposed technique achieves the safe and verifiable computation of the average and variance of the collected vehicular sensory data during the data collecting phase. Concurrently, the author used a permissioned BC to maintain a tamper-proof record of the sensory data collected, ensuring reliable and efficient data sharing [97].

Location

The third component of FC privacy that should be considered is location privacy. The location of nodes transmitting or receiving data must be known only by authorized nodes [192]. Li et al. [198] suggested a collaborative-ride hailing service that preserves privacy using BC-assisted vehicular FC. It anonymously verifies users and only reveals a targeted user if all collaborating service providers are present, with no need for a trusted authority. The authors used a consortium BC to track c-ride data and build smart contracts to connect passengers and drivers. Location authentication, driver screening, and destination matching are all supported via private proximity tests and query processing. They also tweaked Zerocash to enable anonymous payments and fight against double-spending assaults [198].

Kang et al. [83] developed a privacy-preserving pseudonym system with hierarchical architecture. Pseudonyms are created in real-time and supplied to cars. Safe communication methods for privacy preservation are intended for secure and effective pseudonym management. The authors also demonstrated a situation-aware pseudonym shifting game for automobiles that uses context awareness to alter pseudonyms. The suggested architecture enables safe communication and privacy preservation for cars, according to the security analysis [83].

Patwary et al. [165] suggested a distributed location-based device-to-device mutual authentication system for fog devices at the FC layer, without relying on an intermediate third-party system. Using Ethereum smart contracts, they evaluated BC technology to execute the mutual authentication process. Only a few keys are required by the fog devices for authentication. As a result, the suggested approach satisfied security criteria such as data integrity, confidentiality, mutual authentication, and device anonymity. The suggested technique is computationally efficient, according to the performance evaluation. However, due to the location validation procedures conducted, the suggested system needs greater computing overhead in some situations than previous approaches [165].

Insights and discussion

For BC, privacy-preserving strategies based on encryption approaches are evolving, allowing users to become anonymous and have the ability to manage their personal data (e.g., what, whom, and when personal data can be shared in each transaction). Authors proposed several mechanisms to enhance privacy ([174, 209, 218]), to ensure data privacy ([97, 146, 188]), and to enhance location privacy ([83, 165, 192, 198]). The majority of the selected studies under this category reported that BC can enhance the level of privacy, in general, followed by identification, and the least purpose mentioned was to achieve data privacy.

Access control

The tactics or strategies (countermeasures) employed to ensure security goals are referred to as access control [12]. Secure access to data can be ensured using BC in cloud-FC-IoT architecture [186]. This section discusses access control-related purposes including authentication, authorization, and key management.

Authentication

Authentication makes sure users are who they say they are. Malicious nodes, fraudulent communications, and unregistered entities are all targets for authentication techniques [211]. Authentication has been identified as a significant problem in FC [14]. Hewa et al. [52] offer a BC with FC integration security service model that runs on FC. Due to the use of BC, the proposed model ensures privacy and authentication. In comparison to current systems, the suggested model demonstrated a higher degree of security and performance. Secure real-time data on items in transit and supply chains necessitates bandwidth with capacity that the present infrastructure cannot provide. To address this challenge Jangirala et al. [121] proposed LBRAPS which is a new lightweight BC-enabled RFID-based authentication mechanism. Only one-way cryptographic hash, bitwise exclusive-or, and bitwise rotation operations are used in LBRAPS [121]. When a regional fog/cloud demands a lot of verification, it causes traffic problems and delays in the master fog/cloud. Kwon et al. [199] proposed a multi-fog/cloud authentication method based on BC to tackle the problem. To overcome this issue, this system distributes an excessive amount of authentication requests around the fog/cloud region. By unifying dispersed multi-fog/cloud throughout the BC network, it increases authentication times [199].

Yao et al. [75], for distributed vehicular fog services, developed a BC-assisted Lightweight Anonymous Authentication (BLA) approach. BLA can benefit from the following: 1) Implementing a flexible cross-data center authentication system in which a vehicle can choose whether or not to be authenticated while entering a new vehicular fog data center. 2) Establishing anonymity and entrusting vehicle users with the task of maintaining their privacy. 3) It is lightweight due to the lack of interaction between cars and service managers, as well as the elimination of communication between SMs during the authentication process, resulting in a considerable reduction in communication delay. BLA provides these benefits by integrating contemporary cryptography and BC technology uniquely [75]. To establish a secure smart vehicle system, Baker et al. [152] presented a lightweight system that uses BC for authentication. To develop the system, the authors used 5G and federated learning in FC. When compared to the present cloud-based framework, the proposed system showed a high enhancement in security level.

Authorization

The authorization ensures access to a resource only for authenticated users. Authorization is another important aspect of FC security [156]. The BC participant approach guarantees the data necessary to assess the quality of IoT services is reliable. To offer great QoS in highly mobile networks, secure and trustworthy transmission is essential [156].

To solve difficulties related to QoS and data storage, Bouachir et al. [64] suggested industrial cyber-physical systems based on BC with FC integration. Distributed data storage and management over the FC, according to the author, are potential answers to data storage and QoS issues [64]. As an approach to eHealth services, Islam et al. [159] suggested a novel BC with FC integration management system focused on the creation of clustered-based extracted features for the detection of human activities. Bag-of-features, based on Speed-Up Robust Features (SURF), were utilized in the proposed system to pick interest spots for human actions in films. The suggested system’s efficiency and accuracy are improved by using the Error-Correction-Output-Codes (ECOC) method, which allows for classifying multi-class actions [159].

To build confidence in smart apps and the underlying decentralized system, Kochovski et al. [160] looked at several factors that must be evaluated and applied. While certain trust characteristics can be gained through expensive on-BC activities, others can be achieved using less expensive off-BC techniques, such as the usage of data QoS monitoring. To attain good QoS of smart apps, the authors use off-BC QoS monitoring data acquired via a trustless Smart Oracle, as well as a Markov decision-making mechanism that rates the various FC/cloud node providers to pick the best fog node for the AI component of the application’s deployment [160]. Debe et al. [76] proposed a new system for monetizing BC-based services and automating bitcoin payment for services delivered by fog nodes. The suggested method is trustworthy, decentralized, and automated, which enhances QoS and customer satisfaction. The suggested approach governs interactions between FC and devices using the Ethereum BC and its inherent smart contract capabilities [76].

Payment

The incentive and penalty systems utilized by the fog node for BC’s participants are referred to as payment, in this context. Debe et al. [91] proposed a decentralized reverse-bidding method based on BC and smart contracts’ main characteristics. They created a system that allows devices to start the bidding process by requesting services from nearby fog nodes that respond with bid proposals. The suggested method guarantees that all fog nodes on the network compete for the bid fairly and equitably. The automatic payments after the service are included in the bidding procedure. Ethereum smart contracts were used to implement this solution. This method also included a fog node’s reputation system, as well as a penalty for nodes that misbehave [91]. Moreover, Liu et al. [100] proposed distributed BC-inspired energy coins and data coins.

By utilizing the advantages of smart contracts of BC, Jain and Kumar [213] created a fair and trustworthy incentive mechanism that promotes sellers and buyers to transact. Various economic attributes, such as budget balance, personal reasoning, and honesty, are satisfied by this mechanism. The incorporation of the BC and FC precludes the manipulation of trade-related data. The suggested technique was shown to be effective in identifying the winner and pricing model. Shukla et al. [196] demonstrated a BC-based smart energy trading algorithm and a BC with FC integration-based system for P2P energy trading. The proposed algorithm creates a completely trustworthy, low-latency communication network that allows prosumers to trade energy inside their neighborhood, based on the evaluation results. Boualouache et al. [51] developed a monetary reward strategy for 5G-enabled FC-based vehicle location privacy preservation. This solution makes use of a consortium BC in the FC layer as well as smart contracts to assure pseudonym changing procedures and lower vehicle monetary expenses. This scheme provides appropriate monetary cost management and private verification of blocks, according to the evaluations.

Insights and discussion

Because each node in the consortium BC, for example, has access to the data and business norms, the BC’s transaction may be trusted. The BC ledger can now be used to register and exchange nearly anything without the need for a single authority. As a result, a trustworthy and successful network can be initiated. Moreover, by assuring that a fog node is in command of its identification, the immutability of BC gives the necessary reliability and confidence for corporations among nodes. The basic idea is to provide fog nodes identifications that can be verified with BC throughout their entire cycle. A record or timeline is created by a system with an identification, which is managed by a BC. The vast bulk of BCs is open-source, meaning that nodes can see and use their transactions. Users may look up the record of all transactions in the case of Bitcoin thanks to BC transparency. As a result, there will be more openness, which will improve productivity. Bitcoin, for example, is changed when a large majority of network users agree that there is a need for updated code that sounds beneficial. Authors proposed several strategies to ensure trust support ([57, 43]. On retrieval, the data’s integrity may be checked by recalculating its hash value and comparing it to the one that is stored immutably on the BC [43]. Bai et al. [111] proposed a Multiedgechain structure, from the aspect of real-time operation and stability, that supports a big amount of data and improves on-chain data efficiency to provide cross-chain data sharing for diverse BC platforms. Furthermore, a two-stage Stackelberg game tactic was presented, taking into account the risk considerations and user preferences, to maximize the profitability of computing resource scheduling on the Internet of energy [111]. Ismail et al. [145] proposed a framework to enhance data sharing by employing BC methods and data operations to prevent data from altering. IoT may be used to remotely monitor a patient’s status, as well as follow up and provide information to the appropriate authorities, alerting them to potentially harmful circumstances. The data is obtained from the patient, processed in operations, and then saved to communicate trustworthy and reliable information between the caregivers and the patient [145].

Several research initiatives have recently been completed to allow the collaborative platform to create successful collaboration with the manufacturing, design, and consumer perspectives. However, establishing trust and effectively utilizing consumer perspectives remains a difficulty. As a result, Barenji et al. [167] suggested a BC-enabled FC-based collaborative platform to foster triple communication and collaboration in a secure environment across the manufacturing, design, and client sections. Machine learning was utilized to cluster and categorize customer views in the proposed platform, and FC-based integration across subsystems using BC technology is proposed to increase data integrity and security [167]. According to Shahbazi and Byun [212], BC can shift the smart manufacturing on edge computing servers from a cloud-centric to a distributed system FC architecture. In their proposal, the BC technology makes use of data transfer and production system transactions, while the machine learning method allows for enhanced data analysis of a large manufacturing dataset [212]. Rivera et al. [90] proposed a BC framework to offer a trusted cooperation mechanism between edge servers. A permissioned BC approach is being studied in particular to support a trusted design that also offers incentives for collaboration [90].

To accomplish safe data storage and sharing in vehicle edge networks, Kong et al. [157].

Validation

The data transferred from the fog to the cloud will be altered. As a result, the user of an IoT device will never be able to check the accuracy or integrity of data saved in the cloud [14]. When BC is used in conjunction with FC, data validation guarantees that the access token and digital signature (for example, in the smart contract) are valid before the review is stored [138]. Simpson and Quist-Aphetsi [142] suggested a framework that makes it simple to ensure that a patient’s medical information is accessible across multiple healthcare institutions. The usage of a BC ledger allows databases to utilize timestamps to validate and maintain current patient health information in a centralized data cloud [142].

Tian et al. [138] presented a custom-built public auditing technique for data storage that fulfills security and performance requirements. During the proof generation stage, they designed a tag-transforming mechanism based on the bilinear map** technique to translate tags generated by mobile sinks to tags created by fog nodes. This technology not only efficiently preserves identity anonymity but also saves time and money throughout the validation step [138]. Li et al. [65] suggested a carpooling method that supports conditional privacy, destination matching, one-to-many matching, and data auditability utilizing BC with FC integration-based vehicular networks. This method verifies users in a conditionally anonymous manner. Also, it uses one-to-many proximity pairing using a private proximity test and extends it to provide a secret communication key between a client and a driver. A private BC was created to keep track of carpooling records [65].

Insights and discussion

BC can guarantee safe data sharing because due to its distributed and immutable capabilities. Financial firms can watch each transaction in live time thanks to the data stored in BC, enabling them to examine possibly fraud cases. Hence, the BC with FC integration can assist financial firms in preventing fraud and safeguarding their consumers. Additionally, this integration enables service providers to exchange data with other stakeholders while minimizing the risk of data loss. Furthermore, if the data comes from a variety of sources, the need for repeated data analysis may be avoided because each transaction is recorded in the BC. Smart Contracts can be used to govern the data sharing and storage process in BC. On the other hand, to enable large data communications, BC can assure big data training and avoid data breaches. Authors have introduced several strategies to support data management: to secure data storage ([43, 163, 178, 208, 225]), ensure data sharing ([43, 117]. Moreover, multiple access mobile edge computing looks to be an advantageous approach to solve the PoW problems for mobile users in future mobile IoT systems, facilitating BC applications in future mobile IoT systems. Accordingly, ** a quantum computer and individuals and businesses started using it, the algorithm will be broken, rendering nearly all BCs unsafe. There is now a large effort underway to evaluate and standardize post-quantum cryptography primitives [22]. Despite the efforts made to solve the quantum issue (e.g., [233,234,235,236], quantum resilience becomes a serious concern when we construct systems based on BCs that we hope to maintain for many years [238, 239].

Deep learning, along with quicker processors and bigger storage capacities, has cleared the path for modern auditing. Machine learning algorithms, on the other hand, are at the heart of AI and are characterized by their opacity. In this sense, BCs can give auditable trails to show why an AI system made a specific choice and reconcile inconsistencies caused by non-linear usage of many variables and randomization. AI enables a slew of fascinating and creative BC-based applications that might improve the technology’s transparency [22]. The learning process requires a good data sample to create acceptable training data sets. If the adversary is aware of the attack type and has access to the training dataset, the attack type may be readily changed. As a result, understanding the exact nature of an attack to distinguish between desirable and undesired network states is a challenging topic that requires further investigation [109].

The BC structure, which is claimed to be safe and verifiable, may be utilized to make massive data administration easier. Data analyses utilizing the BC structure, on the other hand, entail far very high overhead. Notwithstanding, most cases do not necessitate evaluating all transactions, therefore intermediary or economical supplementary constructs can be developed, increasing overall performance. Despite efforts to introduce big data analysis, traditional big data analysis remains a significant barrier to BC with FC integration [109]. The resources for fog nodes and BC are still limited. Uploading the data to clouds for processing and big data analysis can be a solution, however, this might cause severe latency and privacy issues. Furthermore, anonymized data might make big data analysis difficult to implement, and decrypting data a time-consuming process, resulting in inefficient data analytics. In a nutshell, these new technological developments will have a significant influence on FC performance, making the total integration of BC with FC integration problematic.

Discussion

The body of knowledge on BC with FC integration is relatively scattered. As a result, this research conducted an SLR and presented a holistic explanation of the purposes of this integration. The purpose of the paper was to address two research questions: How do the purposes of blockchain-fog computing integration develop over time? What are the future challenges in integrating blockchain with fog computing? (RQ2). We evaluated all relevant literature in all reputable databases, including IEEE, Elsevier, Springer, MDPI, Google Scholar, Taylor, Sage, ACM, and Emerald, in order to address the research questions. This section offers an overall evaluation, implications for the findings, and limitations of this study.

Security, privacy, access control, trust management, data management, scalability management, and performance were the seven purpose categories that this study identified and discussed. The whole transaction history is contained in the data in the BC, which is hashed to protect the ledger. As a result, BC can make sure that the devices are connected. The combined monitoring of linked fog nodes in the BC system makes data fabrication nearly impossible. Data is therefore dispersed among a large number of connected fog nodes to safeguard BC. This improves the transaction’s security, integrity, and confidentiality.

Additionally, employing BFC-based apps will make it simpler to spot fraudulent activity because if an attacker modifies the data in a block, the block’s hash value will change and the block will become invalid. Therefore, only authorized users may access data without going through extra checks if, for instance, many business units within a firm may participate in a shared BC that offers a degree of access control. Moreover, the immutability of BC provides the required dependability and confidence for companies among nodes by guaranteeing that a fog node is in control of its identity. As a result, there will be more transparency, which will boost productivity and trust. The integrated data analytics capabilities of BC also enable financial institutions to settle cross-border transactions, particularly those involving significant quantities of money, in almost real-time. They can also see how the data is changing in real-time, which enables them to make decisions like transaction banning in real-time.

While a lack of FC resources mostly contributes to the scaling difficulty, BC design and a lack of regulations may lead to security, privacy, and standards difficulties. On the other hand, both BC and FC capacities are impacted by quantum, AI, and big data. These difficulties will affect FC’s performance in each of these cases. Additionally, the lack of standards provides a challenge to the effectiveness of BC with FC integration; hence, future research and industry efforts must concentrate on develo** novel methods and efficient distributed control systems to regulate BC with FC integration. The findings of this study are established in the publicly available literature. Although the use of the Bitcoin platform may have contributed to many challenges and future developments, alternative platforms, such as Ethereum, multichain, and others, should also be looked into. Additionally, since both FC and BC technologies are relatively young, future research and industrial efforts are evolving daily, which makes it challenging to review all data in real-time. The results of this study therefore only applied to the first quarter of 2022. Therefore, future research may use these findings as a foundation and move on from there.

Conclusions

While FC has gained widespread acceptance as a solution to various cloud computing shortages, many concerns remain unresolved. Many of these difficulties can be addressed by combining FC with BC. BC with FC integration seems to provide more secure, scalable, and efficient applications through the combination of BC and FC capabilities. While BC with FC integration seems reasonable, however, there is a need to provide a synthesized knowledge base of their purposes including challenges for future research directions. We addressed this significant requirements and provided a systematic review and synthesis of recent studies, published in the public domain, with a special emphasis on BC with FC integration purposes. We identified seven major themes of BC with FC integration purposes including security, privacy, access control, trust management, data management, scalability, and performance. Within each of these themes, several purposes were also identified and discussed. These themes and underpinning purposes intend to help academics and practitioners to formulate BC with FC integration strategies for their effective adoption of IoT data handling. Moreover, the critical open research problems, impeding the broad use of BC with FC integration, were also identified and reported in this paper. By offering major advances in terms of security, privacy, data management, and trust management, it is anticipated that BC can restructure and revolutionize the future of FC technology. However, BC with FC integration raises several technological issues, including scalability, a lack of standards and regulations, quantum resilience, and AI advancement, which could be further explored in future research studies.