2023 Rani

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Internet of Things 21 (2023) 100672

Contents lists available at ScienceDirect

Internet of Things
journal homepage: www.elsevier.com/locate/iot

Research article

Blockchain-enabled cooperative computing strategy for resource


sharing in fog networks
Shalli Rani a , Divya Gupta b ,∗, Norbert Herencsar c , Gautam Srivastava d,e,f ,∗∗
a Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India
b Department of Computer Science and Engineering, Chandigarh University, Mohali, India
c
Department of Telecommunications, FEEC, Brno University of Technology, Technicka 12, 61600, Czech Republic
d
Department of Mathematics and Computer Science, Brandon University, Brandon, Canada
e
Research Centre of Interneural Computing, China Medical University, Taichung, Taiwan
f
Department of Computer Science and Mathematics, Lebanese American University, Beirut, Labanon

ARTICLE INFO ABSTRACT

Keywords: Several heterogeneous network designs have been developed to fulfill user experience needs,
Fog computing driven by widely growing applications needing large data processing, such as Industry 4.0.
Industry 4.0 Fog computing is one of them since it allows for the efficient integration and exploitation of
Security
ubiquitous computing resources. Willingness and service billing concerns become important
Blockchain
in fog computing situations for computing resource sharing. The present fog systems, on the
other hand, are vulnerable to malicious attacks. The integration of blockchain technology
into the fog computing environment is characterized by successfully allowing consensus in an
untrustworthy environment. However, creating public blockchains requires a lot of processing
power, which can quickly deplete the computational resources of IoT-enabled Smart Machines
(SM) in Industry 4.0. This study suggests partitioning the fog system into fog clusters (FC), with
fog nodes (FN) in each cluster sharing the same access control list (ACL) that is protected by a
public blockchain. Therefore, Blockchain-enabled Resource Sharing inside SM in Fog networks
(B-RSSF), a novel architecture for computing resource sharing in fog networks is proposed.
B-RSSF has specified physical architecture, design principles, and inner workings. To make
full and effective use of ubiquitous computing resources, fog computing’s wireless features
and blockchain technology are tightly integrated. In addition, an enhanced PoW consensus
mechanism for reaching consensus in an untrustworthy fog environment has been discussed.
Finally, the experimental results have been conducted to evaluate the performance of the
proposed scheme over other existing schemes in a fog-based environment.

1. Introduction

Various smart machines (SM) based on Industry 4.0 applications have shown their value, thanks to the increasing development
of Internet of Things (IoT) enabled devices [1]. The sensors embedded in these machines perform sensing, computation, and
communication of data for industrial automation in a wireless communication environment. However, SM inherently acquires
immense power to carry out various functions, which is against the requirements of resource-constrained devices, such as limited

∗ Corresponding author at: Department of Computer Science and Engineering, Chandigarh University, Mohali, India.
∗∗ Corresponding author at: Department of Mathematics and Computer Science, Brandon University, Brandon, Canada.
E-mail addresses: shalli.rani@chitkara.edu.in (S. Rani), divya1907gupta@gmail.com (D. Gupta), herencsn@ieee.org (N. Herencsar),
srivastavag@brandonu.ca (G. Srivastava).

https://doi.org/10.1016/j.iot.2022.100672
Received 14 July 2022; Received in revised form 27 September 2022; Accepted 14 December 2022
Available online 16 December 2022
2542-6605/© 2022 Elsevier B.V. All rights reserved.
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 1. Design of blockchain system.

power and storage availability. Cloud computing has therefore been introduced to alleviate the resource limitations where SM may
perform task computation offloading onto cloud servers and acquire results from the cloud only [2]. Although the utilization of
the cloud will reduce the burden on SM, computation at the distant remote servers will incur high network bandwidth usage and
long transmission delay for getting the results back. Therefore, this paradigm is intrinsically unsuitable for delay-sensitive real-time
Industry 4.0 applications such as the e-healthcare sector and video-on-demand streaming.
To address these challenges of Industry 4.0 communications, a new paradigm of fog computing to mitigate long transmission
delays was proposed as a solution in recent years. The fog computing architecture mainly comprises several fog nodes located
at the edge of the network, each with substantial resources to perform computation, communication, and caching. The core
network facilitates direct communication among these fog nodes. The presence of various fog nodes near end users improves system
performance while offering reduced delays and efficient communication [3].
Nevertheless, in Industry 4.0 applications, the flow of data from different locations follows an open channel, i.e., the Internet,
introducing a high magnitude of threat to security and privacy concerns. Providing security to a large volume of data produced
by various Industry 4.0 applications demands an efficient security mechanism applied at a certain level to support secure end-
to-end communication. To support reduced delay for information processing, the access control list (ACL) gets deployed on fog
nodes, and therefore, the vulnerability to attacks by malicious users is pretty high on the fog computing layer in contrast to the
cloud layer. To ensure access from legitimate users, blockchain technology was first introduced to keep electronic cash systems
safe. Today, public blockchain is the most successful decentralized approach with support to various automatic operations without
any third-party manager and is the best to solve the trust issues in computing resource sharing. Smart contracts and consensus
mechanisms are the fundamental practices of blockchain to ensure security and trust in a nonreliable environment with additional
promise for no data tempering [4]. The traditional blockchain network as shown in Fig. 1 presents a design of blockchain solution or
system showing interactions of different components including Smart contracts, storage or cloud and IoT-enabled clients. However,
maintaining a single blockchain at the fog computing layer to ensure synchronization and accuracy of ACL during resource sharing
is not practical. The promising solution would be dividing the entire fog computing system into several fog clusters (FC), where
each cluster maintains a blockchain [5].
These small P2P networks, i.e., FC, will not only reduce the computation power requirement for the generation of blockchain in
every cluster to a great extent but will also support reduced latency during resource sharing with increased trust security.
This article proposes the architecture of blockchain-enabled fog clusters (B-FC) to enable distributed resource sharing among
various SM in a fog computing environment, named Blockchain-enabled Resource Sharing inside SM in Fog Networks (B-RSSF). The
key components of the proposed network architecture are represented in Fig. 2. The rewards provided by smart contracts motivate all
SM to compete in the resource-sharing process. However, fog nodes in B-FC are not rewarded after they finish the resource-sharing
process. Therefore, the relation among fog nodes in B-FC is cooperative rather than competitive.

2
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 2. Representation of key components in the network architecture.

The main contributions of this study are:

• B-FC architecture to support resource sharing: The proposed B-RSSF is a decentralized approach to offer efficient resource
sharing among SM with no requirement of any third-party manager. All the fog nodes in one B-FC jointly maintain a blockchain
to increase the difficulty of data tempering, thus ensuring a highly secure system in a resilient wireless communication
environment without any third-party supervision. We also design a reward procedure to increase the willingness of the SM to
participate in the resource-sharing process by facilitating their accounts with fixed credits for each shared resource through
the blockchain system.
• B-FC architecture suitable for resilient wireless communication environments: The design architecture of the proposed B-
RSSF has been carefully designed to meet the requirements of a resilient wireless communication environment. The blockchain
system embedded inside the fog computing layer for the synchronization of ACL ensures secure wireless communication in
this architecture. We also consider key features of wireless network communications such as the transmission capability of
wireless links and the credibility by maintaining the reputation record at each node.

The rest of the article is organized as follows: Section 2 presents related work in this domain. The background knowledge of
blockchain networks and blockchain in industry 4.0 has been presented in Section 3. In Section 4, the blockchain-enabled fog
cluster is introduced. Furthermore, the B-RSSF design principle along with its working is presented in Sections 5 and 6, respectively.
Following this, Section 7 provides a discussion on a consensus mechanism in the proposed approach. The experimental results of
the proposed scheme have been provided in Section 8. Finally, Section 9 concludes this study.

2. Related work

Gu et al. presented CrowdChain [6], a fog-assisted blockchain-based crowdsensing framework in which blockchain is used to
execute secure payments and rewards, as well as to store identities in a fog-based chain. The work in [7] suggested partitioning
the fog layer into two virtual fog clusters. One of them functions as an intermediate layer between the IoT and the cloud. The
other cluster is dedicated to blockchain-related work. The delay effect produced by the installation of a PoW-based blockchain was
extensively explored in this study. The work in [8] presented a paradigm in which blockchain is installed in an IoT-Fog-Cloud context
to store data about the fog node’s contributing resources. In this approach, the generated blockchain gives a valid and credible fog
node evaluation index. For Industrial Internet of Things (IIoT) applications, the work in [9] presented the usage of blockchain for the

3
S. Rani et al. Internet of Things 21 (2023) 100672

Smart Industry. By leveraging Ethereum and Proof-of-Authority (PoA), the idea was to establish a marketplace for smart Industry-
related applications in support of different technologies such as fog computing, blockchain and virtual reality. Blockchain was
installed to handle various trusted payment transactions, consumer and supplier authentication, and data storage-related activities.
The work in [10] is based on the study of the integration of blockchain with fog computing for smart factories. To implement
this, different fog and cloud nodes act as blockchain nodes in the proposed cloud-fog architecture. The main purpose of such nodes
was to keep a record of transactions being carried out in complete architecture. The work in [11] focusses on using a blockchain
architecture and fog computing aspects to develop a privacy preserving framework for Internet of Things driven smart cities. Based
on their work, a novel methodology is presented to address current limitations in security and privacy in smart city environments.
Further, in the Industry 4.0 scenario, the deployment of blockchain for supply chain management system was proposed [12]. The
introduction of blockchain in this framework replaced conventional databases by storing the transactions between IoT and fog, and
decisions performed by the cloud. Although all of these works focused on providing security in different aspects while considering
reduced complexity. Still, computational complexities are very high in small-scale fog networks. In addition, creating blockchains
requires a lot of processing power, which can quickly deplete the computational resources of IoT-enabled Smart Machines (SM) in
Industry 4.0. This study suggests partitioning the fog system into fog clusters (FC), with fog nodes (FN) in each cluster sharing the
same access control list (ACL) that is protected by a blockchain, thereby offering reduced computational complexity and burden at
fog nodes.

3. Background

3.1. Blockchain: An overview

Blockchain is a distributed, immutable ledger that significantly improves the accessibility of recording transactions and managing
assets in a corporate network. Virtually everything of value may be recorded and traded on a blockchain network, reducing risk and
increasing efficiency for all parties. The technical foundation that enables applications to access ledger and smart contract services
is known as a blockchain network. The primary function of smart contracts is to create transactions, which are subsequently sent
to every peer node in the network and permanently stored on each node’s copy of the ledger. Examples of app users include end
consumers who use client applications or blockchain network administrators. Different methods can be used to create a blockchain
network. They may be permissioned, public, private, or built by a consortium of users [13].

3.1.1. Public blockchain


Anybody is welcome to join and take part in the essential operations of the blockchain network on a public blockchain. The
self-governed, decentralized aspect that is frequently praised when discussing blockchain is made possible by the fact that anybody
may read, publish, and audit the current actions on a public blockchain network. The fact that public blockchains can act as the
foundation for almost any decentralized solution makes them incredibly important. Additionally, a secured public blockchain is
protected from data breaches, hacking attempts, and other cybersecurity problems by the large number of network users that join
it. A blockchain is safer the more participants it has.

3.1.2. Private blockchain


Permissioned blockchains that are managed by a single entity are referred to as private blockchains or managed blockchains. In
a private blockchain, a node’s eligibility is decided by the central authority. Additionally, the central authority does not necessarily
give every node the same permissions to carry out tasks. Private blockchains are only partially decentralized, though, because they
are not accessible to the general public. Private blockchains lack many of the beneficial features of permissionless systems despite
being specifically created for business applications because they are not as extensively used. Instead, they are designed to carry out
particular activities and responsibilities [14]. Private blockchains are vulnerable to data breaches and other security risks in this
regard. This is so that, if a consensus mechanism exists, a consensus about transactions and data can be reached by a finite number
of validators.

3.1.3. Permissioned blockchain


Permissioned blockchains combine public and private blockchains and offer a wide range of customization options.

3.1.4. Consortium blockchain


Contrary to private blockchains, consortium blockchains are permissioned blockchains that are managed by a group of
organizations rather than a single one. Due to this greater decentralization than private blockchains, consortium blockchains have
higher levels of security. Constructing consortiums, on the other hand, can be challenging because it calls for cooperation between
several businesses, which raises logistical concerns and the possibility of antitrust breaches.

3.2. Blockchain in Industry 4.0

Numerous Industry 4.0-based applications have been implemented globally as a result of the widespread usage of the Internet
and related technologies, in which sensors and actuators perceive, process, and transfer data for industrial automation. Moreover,

4
S. Rani et al. Internet of Things 21 (2023) 100672

the flow of data from different locations follows an open channel, therefore risks to security and privacy have multiplied many-fold
as well. Maintaining confidentiality, privacy, and integrity, however, becomes a significant concern in Industry 4.0 because of the
numerous data exchanges that take place over the Internet. Additionally, polls by several organizations indicate that 33 billion
people’s records will have been exploited by 2023, up from 12 billion in 2018 and roughly 60 million people harmed by identity
theft. Current industry solutions use a centralized client–server architecture, where the centralized authority controls all privileges, to
mitigate the aforementioned concerns. However, the entire system may fail if the centralized authority were to be undermined. Data
Encryption Standard (DES), Advanced Encryption Standard (AES), and its derivatives are examples of traditional security techniques.
Blockchain technology can manage a variety of security assaults as it can do away with the requirement for a centralized authority
to carry out various processes.

4. Blockchain-enabled fog cluster

The generation of blockchain in most of its applications consumes a lot of computational power and storage. To address this
challenge of high resource usage, we propose a blockchain-enabled fog cluster (B-FC). The high power consumption for generating
a blockchain in a complete fog network has now been reduced by allowing the generation of a blockchain inside each B-FC. The
generation of blockchain in small-scaled B-FC would offer better resource utilization compared to large-scale fog networks. In each
B-FC, the SM covered in its range can communicate directly with each other following a Device-to-Device (D2D) communication
technology. In addition, B-FC can provide computing power to SM in its coverage range. The main feature of fog nodes is to help
SM to offload their computationally intensive tasks.

4.1. Blockchain in B-FC

In each B-FC, the fog nodes (FN) storing blockchain and ACL are denoted as B-FN. Each block in the blockchain records the
decision of SM (either acceptance or rejection) for resource sharing and changes in ACL during the previous period. Any update
in ACL or any resource sharing request has been marked as one transaction. In addition, the reputation record of all SM is also
recorded within each block in the fog node’s blockchain. The reputation of each SM includes its computational capacity and wireless
communication channel quality. The B-FN in one B-FC work in cooperation and try to respond to SM in the least time to offer
reduced communication delay. On the other hand, SM competes among themselves to satisfy the resource-sharing request following
a competitive reward-based approach. The credit gets rewarded into the account of provider SM after a successful response.

5. The design principle of B-RSSF

Based on the conventional blockchain workflow design, Fig. 3 represents the specific design principle for our proposed B-RSSF
that considers computing resource processing requests in a blockchain-enabled fog computing environment for a resilient wireless
communication environment. The task flow between requester and provider is the same as the process of computing resource sharing.
In our design principle, any requester SM do not directly offload tasks for computing resource sharing to B-FN in their range but
first, attempts to process these requests within a nearby SM. This promotes low latency D2D communication among various SM for
processing resource-sharing requests. Due to the rewards associated with the provider, the SM with enough computational capacity
competes among themselves and hence increased willingness for participation. Any request–response flow within SM, SM to B-FN
or any update is recorded as a transaction in the blocks present in B-FN [15]. B-FN computes the block hash (BH) for the block
by randomly applying different nonce values. The accepted BH is then passed to all available B-FNs inside the current B-FC for
verification. On successful verification of the BH after reaching a network consensus, a new block gets added to the blockchain at
each B-FN in the current B-FC and the provider of computation resource sharing gets rewarded with fixed credits in its account.

6. Working with B-RSSF: Example case

Normally, the requester starts the resource sharing process by requesting a certain computing resource, while providing some
information for this computing resource sharing [16]. The SM in the proximity of the requester will compete among themselves
to become a provider after validating the authenticity of the requester and request. The SM accepting the request for computing
resource sharing will send the request processing information, which mainly contains its computing capacity and channel quality.
The requester chooses one provider for computing resource sharing among all competitors based on the received request processing
information and the node’s reputation record. The communication between the requester and the provider begins after offloading
the computing data to the provider by the requester. Meanwhile, the credit value is deducted from the account of the requester. On
successful completion of the computation task, the results are sent back to the requester by the provider, which is further sent to
the connected B-FN for verification. The results are validated only after all B-FN in the current B-FC reach the network consensus
to accept the BH, which is generated for the block having all these request–response message flows. After validation, the account
balance of the provider is increased to the predetermined credit value, and the computing resource-sharing process finishes. The
complete working of the proposed B-RSSF is illustrated in Fig. 4.
As discussed below, there are four phases for a successful computing resource sharing process request phase, competition phase,
resource sharing phase, and validation phase.

5
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 3. Design principle of B-RSSF.

Request Phase: The requester 𝑆𝑀1 initiates the process by broadcasting request information to all nearby SM which includes credit
value for computing resource sharing, the size of the request, and channel quality.

Competition Phase: While receiving the request by 𝑆𝑀2 …𝑆𝑀𝑝 …𝑆𝑀𝑛 , they first authenticate the credentials of the requester and
request and based on the request information decides on whether to accept/reject the request. All receivers accepting the request
will compete among themselves by sending their requests processing information such as channel quality, computing capability,
and credit value to the 𝑆𝑀1 . Based on the received request information and considering all factors, finally, 𝑆𝑀𝑝 has been selected
as the provider.

Resource Sharing Phase: 𝑆𝑀1 offloads the computing data to 𝑆𝑀𝑝 and the corresponding credit amount is taken from the account
of 𝑆𝑀1 as the computing resource sharing fee. The 𝑆𝑀𝑝 provides the computing results within a stipulated period to 𝑆𝑀1 which
are further validated for their correctness.

Validation Phase: The validation of the received results is performed at the B-FC level by all B-FN. The B-FN connected to 𝑆𝑀1
generates a block hash for the block. After reaching a consensus by all the B-FN for the acceptance of this computed BH, only then
are the results sent back to 𝑆𝑀1 . The computing resource sharing process finishes after adding the corresponding credit value into
the account of 𝑆𝑀𝑝 .

7. Consensus mechanism in B-RSSF

The consensus is the core of blockchain technology to ensure validation in decentralization-based security systems. Generally,
PoX-based consensus mechanisms (such as PoW, PoS, etc.) are most appropriate for blockchain formation, while ensuring fairness
between different nodes along with the decentralized approach [17]. The Proof-of-Work (PoW) consensus mechanism is best suited
for B-RSSF as it defines a complex computing calculation, also known as mining, which should be performed to create a new
group of less trustworthy transactions on a distributed ledger blockchain. Mining is important due to two features (1) to verify the
validity of the transaction and (2) to reward the miners due to the performed previous task. To exhibit better memory consumption,
energy utilization, and convergence time, we have applied the PoW consensus mechanism for fog computing using statistical
likelihood maximization, and polynomial matrix factorization [18]. This enhanced PoW consensus is highly acceptable for cloud or
fog environments with dynamic and resource-constrained devices.

6
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 4. Working with B-RSSF.

During the initial phase of the enhanced PoW, a large number of resources and time is consumed as the system would be
gathering all required information from the environment. However, there would be a gradual decline in the consumption of both
resources and time for further executions. We considered some assumptions while implementing this PoW, such as (1) difficulty
level only based on the number of zeros at the starting of the block hash. (2) The Nonce values are randomly chosen from 256
bits to 2048 bits. (3) SHA-512 is utilized for hashing operation. The basic procedure of reaching consensus by different B-FN in
one B-FC based on PoW in B-RSSF is represented in Fig. 5. The computed result present at 𝑆𝑀1 would be broadcasted to B-FN of
corresponding B-FC for verification. Once B-FN (B-FN1, B-FN2, B-FN3, B-FN4) receive the request, they work in cooperation to get
the acceptable block hash value for the block of transactions. To compute an acceptable block hash within the minimum possible
time, the set of nonce values gets divided into the number of subsets, which is equal to the number of B-FN in B-FC. Each B-FN
performs computations on its own provided subset of nonce values to generate an acceptable BH. Any B-FN successfully computing an
acceptable BH broadcast nonce to all other B-FN for verification. Each B-FN verifies the computed BH to reach a network consensus
based on PoW. The acceptable BH is then broadcasted to each B-FN in the current B-FC to add a new block to their blockchain.
The verified result is sent back to 𝑆𝑀1 to ensure the correctness of the computing task performed by 𝑆𝑀𝑝 . Differently, in B-RSSF,
due to limited wireless communication connections and a highly dynamic environment, sometimes B-FN computing acceptable BH
first may not broadcast to the network at first, and hence not be able to get its reward. Therefore, the B-FN in B-FC is encouraged
to work in cooperation rather than competition. In addition, the collaboration among B-FN will reduce the resources consumed at
each node due to the subset of nonces present for computation, which earlier was a complete set for each B-FN in a competitive

7
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 5. Reaching consensus in B-RSSF.

Fig. 6. Computation time with different SMs.

approach. The proposed blockchain solution include built-in security features. It is founded on cryptographic, decentralized, and
consensus concepts that guarantee the integrity of transactions. The data is organized into blocks in the majority of blockchains
or distributed ledger technologies (DLT), and each block contains a transaction or collection of transactions. In a cryptographic
chain, each new block is connected to all the blocks that came before it in a way that makes tampering with it nearly impossible. A
consensus mechanism verifies and accepts each transaction contained within the blocks, ensuring that each transaction is accurate
and true. Decentralization is made possible by blockchain technology by allowing members of a dispersed network to participate.
No single point of failure exists, and one individual user cannot alter the record of transactions. This ensures security in terms of
confidentiality, data integrity, data tempering, and from Sybil attacks.

8
S. Rani et al. Internet of Things 21 (2023) 100672

Fig. 7. Computation time with different computational power.

Table 1
Computation time with different average SMs.
No. of SMs Proposed EDA RDA
15 11 13.4 12.3
20 10.5 12.8 11.8
25 9.7 11.6 10.7
30 9.2 11.2 10.4
35 8.2 9.7 8.95

Table 2
Computation time with different computational power.
Average computing power Proposed EDA RDA
600 8.4 10.3 9.2
700 7.8 9.2 8.4
800 6.3 8 7.2
900 5.9 7.4 6.7
1000 5.1 6.2 5.8

8. Simulation results

In this simulation, we determine the computation time to obtain the first valid BH from a B-FN with 15 SMs. The computing
power of each SM is distributed normally along the interval N(500, 200) (hash/s). N(2000, 1000) is a normal distribution for the
network delay (milliseconds). The likelihood of having acceptable BH is = 10−4 . Equal division allocation (EDA) and ratio-oriented
allocation are two currently used allocation systems and their respective performances are compared with the proposed scheme. All
devices having processing power are given an equal share of the computing set via EDA. The amount of available computing power
as well as the network delay are not taken into account while allocating computing sets using EDA.
The computing sets are distributed by RDA according to the available computing power of the SMs, thus the SM with more
computing power receives a larger computing set. Fig. 6 displays the computation time with various numbers of MDs covered by a
single BFN. The experimental values retrieved after evaluation have been presented in Table 1. The overall amount of computational
power that is available can be increased by adding more SMs to a B-FN’s coverage. As a result, all three methods can acquire the
first acceptable BH more quickly.
This is so that each SM may find a suitable BH in its associated computing set in less time as the number of MDs grows and the
computing set allotted to each MD gets smaller. As a result, the network delay is contributing heavily to the entire computation
time. Moreover, the proposed scheme can jointly take into account the network delay and the number of computing sets while
allocating computing sets to shorten the processing time required to get the first acceptable BH, While EDA and RDA are unable to
do so.
The available computational power of SMs is another factor that affects how quickly the first acceptable BH may be acquired.
Fig. 7 displays the length of time required to compute the first acceptable BH using various techniques with various average computer
power. The experimental values retrieved after evaluation have been presented in Table 2. Because WEAR takes into account both
the available computing power and the network latency when allocating the computing set, the computing time it takes to acquire
the first acceptable BH is shorter than that of RDA and EDA with the same amount of computing power. RDA more effectively cuts
down on computing time than EDA by allocating large computing sets to MDs with powerful computers.

9
S. Rani et al. Internet of Things 21 (2023) 100672

9. Conclusion

This article presents a novel B-RSSF architecture to facilitate computing resource sharing in fog networks by integrating wireless
communication and blockchain technology for Industry 4.0 requirements. Due to the untrustworthy characteristic of fog computing
networks, the concerns related to the willingness and billing of services in computing resource sharing are explored in depth. The
B-SSF architecture is explained, in which the whole fog network is separated into multiple clusters to decrease the computing
power requirements for blockchain creation. The design principle and working of B-RSSF based on blockchain technology are all
developed and depicted. Furthermore, a detailed look at essential B-RSSF technology, such as consensus mechanisms, especially
when considering the wireless features of fog computing and blockchain technology has been provided. The experimental results
obtained after simulation have proved the efficiency of the proposed scheme over existing fog-based blockchain solutions. However,
the computational overhead involved in dividing the whole fog network into various clusters and assigning SMs with each fog
node has been ignored in this study. In future work, we would design a blockchain-based security solution in a fog environment
considering above mentioned computational overheads in the system’s overall performance.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared
to influence the work reported in this paper.

Data availability

Data will be made available on request.

References

[1] Y. Lu, Industry 4.0: A survey on technologies, applications and open research issues, J. Ind. Inf. Integr. 6 (2017) 1–10, http://dx.doi.org/10.1016/j.jii.
2017.04.005.
[2] S. Iqbal, A.W. Malik, A.U. Rahman, R.M. Noor, Blockchain-based reputation management for task offloading in micro-level vehicular fog network, IEEE
Access 8 (2020) 52968–52980, http://dx.doi.org/10.1109/ACCESS.2020.2979248.
[3] M. Mukherjee, L. Shu, D. Wang, Survey of fog computing: Fundamental, network applications, and research challenges, IEEE Commun. Surv. Tutor. 20
(3) (2018) 1826–1857, http://dx.doi.org/10.1109/COMST.2018.2814571.
[4] V. Gatteschi, F. Lamberti, C. Demartini, C. Pranteda, V. Santamaria, To blockchain or not to blockchain: That is the question, IT Prof. 20 (2) (2018)
62–74, http://dx.doi.org/10.1109/MITP.2018.021921652.
[5] D. Wu, N. Ansari, A cooperative computing strategy for blockchain-secured fog computing, IEEE Internet Things J. 7 (7) (2020) 6603–6609, http:
//dx.doi.org/10.1109/JIOT.2020.2974231.
[6] X. Gu, J. Peng, W. Yu, Y. Cheng, F. Jiang, X. Zhang, Z. Huang, L. Cai, Using blockchain to enhance the security of fog-assisted crowdsensing systems, in:
2019 IEEE 28th International Symposium on Industrial Electronics, ISIE, IEEE, 2019, pp. 1859–1864, http://dx.doi.org/10.1109/ISIE.2019.8781332.
[7] R.A. Memon, J.P. Li, M.I. Nazeer, A.N. Khan, J. Ahmed, DualFog-IoT: Additional fog layer for solving blockchain integration problem in internet of things,
IEEE Access 7 (2019) 169073–169093, http://dx.doi.org/10.1109/ACCESS.2019.2952472.
[8] H. Wang, L. Wang, Z. Zhou, X. Tao, G. Pau, F. Arena, Blockchain-based resource allocation model in fog computing, Appl. Sci. 9 (24) (2019) 5538,
http://dx.doi.org/10.3390/app9245538.
[9] A. Seitz, D. Henze, D. Miehle, B. Bruegge, J. Nickles, M. Sauer, Fog computing as enabler for blockchain-based IIoT app marketplaces - A case study,
in: 2018 Fifth International Conference on Internet of Things: Systems, Management and Security, IEEE, 2018, pp. 182–188, http://dx.doi.org/10.1109/
IoTSMS.2018.8554484.
[10] S.-H. Jang, J. Guejong, J. Jeong, B. Sangmin, Fog computing architecture based blockchain for industrial IoT, in: International Conference on Computational
Science, Springer, 2019, pp. 593–606, http://dx.doi.org/10.1007/978-3-030-22744-9_46.
[11] P. Kumar, R. Kumar, G. Srivastava, G.P. Gupta, R. Tripathi, T.R. Gadekallu, N.N. Xiong, PPSF: a privacy-preserving and secure framework using
blockchain-based machine-learning for IoT-driven smart cities, IEEE Trans. Netw. Sci. Eng. 8 (3) (2021) 2326–2341.
[12] E.N. Lallas, A. Xenakis, G. Stamoulis, A generic framework for a peer to peer blockchain based fog architecture in industrial automation, in: 2019 4th
South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference (SEEDA-CECNSM), IEEE, 2019, pp. 1–5,
http://dx.doi.org/10.1109/SEEDA-CECNSM.2019.8908360.
[13] N. Nizamuddin, H.R. Hasan, K. Salah, IPFS-blockchain-based authenticity of online publications, in: International Conference on Blockchain, Springer,
2018, pp. 199–212, http://dx.doi.org/10.1007/978-3-319-94478-4_14.
[14] I. Yaqoob, K. Salah, M. Uddin, R. Jayaraman, M. Omar, M. Imran, Blockchain for digital twins: Recent advances and future research challenges, IEEE
Netw. 34 (5) (2020) 290–298, http://dx.doi.org/10.1109/MNET.001.1900661.
[15] O. Novo, Blockchain meets IoT: An architecture for scalable access management in IoT, IEEE Internet Things J. 5 (2) (2018) 1184–1195, http:
//dx.doi.org/10.1109/JIOT.2018.2812239.
[16] Y. Gao, W. Wu, P. Si, Z. Yang, F.R. Yu, B-ReSt: Blockchain-enabled resource sharing and transactions in fog computing, IEEE Wirel. Commun. 28 (2)
(2021) 172–180, http://dx.doi.org/10.1109/MWC.001.2000102.
[17] L.M. Bach, B. Mihaljevic, M. Zagar, Comparative analysis of blockchain consensus algorithms, in: 2018 41st International Convention on Information and
Communication Technology, Electronics and Microelectronics, MIPRO, IEEE, 2018, pp. 1545–1550, http://dx.doi.org/10.23919/MIPRO.2018.8400278.
[18] G. Kumar, R. Saha, M.K. Rai, R. Thomas, T.-H. Kim, Proof-of-work consensus approach in blockchain technology for cloud and fog computing using
maximization-factorization statistics, IEEE Internet Things J. 6 (4) (2019) 6835–6842, http://dx.doi.org/10.1109/JIOT.2019.2911969.

10

You might also like