This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEE INTERNET OF THINGS JOURNAL

Blockchain Security Using Merkle Hash Zero

Correlation Distinguisher for the IoT in Smart Cities
Rizwan Patan, Member, IEEE, Ramachandran Manikandan, Ramalingam Parameshwaran,
Sivanesan Perumal, Mahmoud Daneshmand, Senior Life Member, IEEE, Amir H. Gandomi, Senior Member, IEEE

Abstract— Internet of Things (IoT) data is one of the most

important assets in business models for offering various I. INTRODUCTION
ubiquitous and brilliant services. The IoT is provided with the The Internet of Things (IoT) offers a platform to collect, share,
advantage of susceptibility that cybercriminals and other and forward information collected from smart devices daily. IoT
malicious users. Even though smart cities are intended to extend is a key enabler of smart cities, and IoT sensing devices are an
productivity and efficiency, residents and authorities face risks important component for connecting communication and
when they avoid cybersecurity. The conventional blockchain networking devices within cities. Recently, IoT networks and
methods were introduced to ensure the secure management and embedded devices have significantly increased, which
examination of the smart city big data. But, the blockchains are unfortunately presents new dimensions of different threats
found to have computationally high costs, and failed to improve related to security and privacy.
the security, not adequate resource-constrained IoT devices Blockchain technology is the fundamental technology of
have been designated for smart cities. In order to address these Bitcoin, a kind of co-trust smart book technology dependent on
issues, the proposed novel blockchain model called Blockchain the principle of cryptography. Thus far, several methods have
Secured Merkle Hash Zero Correlation Distinguisher (BSMH- applied blockchain to address the challenges of IoT security. For
ZCD) is suitable for IoT devices within the cloud infrastructure. instance, smart cities integrate IoT with nearly every facet of
The objective of the BSMH-ZCD method is to enhance security life, ranging from home security and education systems to
and reduce the run time and computational overhead. Initially, health and personal devices, and will become more predominant
the Merkle Hash tree is used to create the hash value with every with the evolutions of such technology.
transaction. Next, the Zero Correlation Distinguisher is applied The novel privacy-preserving blockchain, called the TrustChain
to perform the data encryption and decryption operation for the approach, was first introduced for developing secure and trusted
ARX block for obtaining proficient secure data access in the IoT data management in smart cities [1]. This research model
devices. Experimental assessment of the proposed BSMH-ZCD explains how TrustChain is employed in the edge computing
method and existing methods are carried out by using the taxi environment with various levels of considerations to eliminate
driver dataset and Novel Corona Virus2019 Dataset with computational delays, and privacy analyses were performed to
different factors such as running time, computational evaluate centralized processing for securing IoT network
complexity, and security with respect to a number of blocks and resources. In the layers of edge computing, small sensors can be
executions. By using the taxi driver dataset, the experimental found at the bottom of the hierarchy, followed by nodes with
results reveal that the BSMH-ZCD method performs better with comparatively higher processing power and storage in the
a 19% improvement in security, 20% reduction of middle, then the large data centers that represent nodes in the
computational complexity, and 29% faster running time for IoT top cloud layer.
compared to existing works. TrustChain is a combination of trust and Byzantine Fault
Keywords—Internet of Things, Smart Cities, Blockchain, Tolerance (BFT), which can be easily established at the end of
Cloud infrastructure, Merkle Tree, Hash Zero Correlation the edge computing hierarchy. Furthermore, a lightweight
Distinguisher protocol introduced in TrustChain enables the efficient
processing of blocks, thereby saving more energy compared to
P. Rizwan is with the Decentralized Science Lab (dSL), traditional blockchain technologies. However, the security was
Department of Software Engineering & Game Development, not enhanced when dealing with delicate personal information.
College of Computing and Software Engineering, Kennesaw To address this existing issue, the proposed Merkle Hash Zero
State University, 680 Arntson Drive, Marietta, GA 30060, USA Correlation Distinguisher (MH-ZCD) is introduced with Add
(e-mail: [email protected]). Rotate XOR block ciphers to increase the security when dealing
Ramachandran Manikandan and Sivanesan Perumal are with the with managing personal information in smart city management
School of Computing, SASTRA Deemed University, Tamil
services.
Nadu, India. (e-mails: [email protected] and
[email protected]). A Self-Redactable Blockchain (SRB) has been presented to
Parameshwaran Ramalingam is with the Department of ECE, provide a secure, acceptably efficient mechanism for IoT
National Institute of Technology, Tiruchirappalli, India (e-mail: devices [2]. With the help of chameleon hash, a redactable
[email protected]) blockchain was obtained. The block content can be re-written
M. Daneshmand is the advisory chair of IEEE IoT and with the without causing major hard forks, while the redacted block hash
School of Business, Stevens Institute of Technology, Hoboken, function remains constant. This produces disincentives to utilize
NJ 07030, USA, (e-mail: [email protected]). redactable blockchain for organizing big data or any data-driven
A.H. Gandomi (corresponding author) is with the Faculty of business mainly because of the ineffective chain reaction. This
Engineering & Information Systems, University of Technology
is where the SRB is introduced to support the intelligent
Sydney, Australia (e-mail: [email protected]).

2327-4662©2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

execution of the chain reaction. First, a Revocable Chameleon data in IoT. Blockchain is one of the promising technologies to
Hash (RCH) was designed to identify the collision without any address IoT issues. Several existing Blockchain-based methods
cooperation. Periodical expiration was also conducted to were developed to enhance security. But, the conventional
prevent any abuse of redaction power. But, the computational Blockchain-based systems for IoT fail to preserve the intelligent
complexity was not reduced by using a consensus mechanism. execution of the chain reaction. However, the computational
In order to overcome the issue. In this work, a Merkle Hash Tree complexity and running time were not minimized. The packet
is introduced in the proposed BSMH-ZCD model. Transactions delivery ratio was not minimized in traditional methods. It failed
in the blockchain via hashes are transmitted publicly in the to obtain information security. To solve the above-mentioned
blockchain network and include supplementary information existing problem, the novel BSMH-ZCD model with the new
about the sender and recipient. The proposed model also uses features of Merkle Hash Zero Correlation Distinguisher (MH-
data encryption via Correlation Distinguisher and Add Rotate ZCD) is introduced in this work. Contrary to existing work, the
XOR block ciphers. Here, double encryption is said to be BSMH-ZCD model aims to achieve higher security, and
ensured by applying the ARX block ciphers and then encrypting minimize the run time and computational overhead.
the symmetric key using the Zero Correlation Distinguisher. Specifically, Merkle Hash Tree (MHT) is applied to minimize
Security can then be attained with minimal computational the running time, while the Zero Correlation Distinguisher
complexity for IoT devices involving big data by utilizing both (ZCD) is used to obtain a higher security level.
techniques.
With the rapid advancement of this digital age, a smart city Research Contribution:
requires effective and sustainable solutions in different sectors. The proposed model research contributions are summarized as
Previous work describes the role played by big data, IoT, and follows:
the necessity of requiring smart city construction utilizing • The main contribution of the proposed BSMH-ZCD model
blockchain technology [3]. The corresponding features were is developed for improving security with lesser run time
analyzed, and final solutions were provided by introducing the and computational overhead.
security aspects, equipment enhancements, and so on. The • The novelty of Merkle Hash Tree (MHT) is applied in the
recent advancements in wireless communications have resulted blockchain framework to permit the hash of the block to
in a wide variety of devices being connected to the Internet, identify the chains of before generated blocks. Then, the
forming the IoT. Routing that establishes a communication path traversal of the entire tree is avoided with aid of hashing
between the IoT source and a destination device is one of the with minimum running time.
major problems currently faced. A Blockchain-based • The Merkle Hash Tree (MHT) and Zero Correlation
Contractual Routing Protocol for a network involving large Distinguisher (ZCD) are utilized in the proposed BSMH-
numbers of un-trusted IoT devices was designed in [4]. ZCD model with lesser computational overhead. Initially,
However, the packet delivery ratio was found to be MHT is applied to create the hash value for each
comparatively less with existing techniques. Considering the transaction. Then, ZVD is used to perform the data
packet delivery ratio, the development of blockchain encryption and decryption operation for the blockchain
technologies requires newer tools for private data management, based on the hash value with minimum computational
therefore ensuring an increased level of security and privacy to complexity.
individual users [5]. To combat such pollution, smart cities have • The innovation of ZCD is employed the proposed BSMH-
become relatively popular over the past few years. The word ZCD model to obtain a higher security level in smart cities.
“smart” is associated with advantages on certain levels of It is designed with two portions of hash values such as left
innovative applications to ensure smooth communal sharing [6] half and right half. It includes the addition, rotation, and
with respect to three dimensions, i.e. humans, technology, and XOR operations for block data encryption/decryption in
organizations. IoT devices to efficiently minimize the attack complexity
Analysis of the resource requirements for DIDs on IoT devices and enhance security in smart cities.
shows that implementing a blockchain on the IoT is rather The rest of this work is organized as follows. Section 2 presents
complex due to resource constraints [7]. To address this issue, a related works in terms of the novel findings and shortcomings
local peer network was used to restrict the frequency of related to this study. Section 3 introduces the problem
transactions entering the global blockchain, utilizing a scalable statement. Section 4 explains the flow diagram along with the
local ledger [8]. Furthermore, the peer validations at both the algorithm for the proposed Merkle Hash Zero Correlation
local and global levels were found to not be compromised. New Distinguisher (MH-ZCD) method for ARX block ciphers.
IoT-based blockchain technology gives up the memory Section 5 includes a comprehensive evaluation of the proposed
constraint and provides a centralized system, in addition to the method’s performance, along with the dataset descriptions.
provisioning of both security and privacy aspects [9]. A survey Section 6 presents the mathematical evaluations and a
of the blockchain-based mechanisms used to provide security comparison of our model to state-of-the-art works. Section 7
for the Internet of Things indicates that the data can be concludes the paper.
safeguarded from data tampering [10]. However, a large amount
of energy is consumed in verifying each block, preventing its II. RELATED WORKS
usage with resource-constrained IoT applications. Another secured mechanism using a blockchain-based
mechanism was presented in [11], whereby validation via edge
Motivation: service providers was achieved by provider identification and
The talented IoT technology is a significant part to ensure the service verification to remove un-trusted services. Smart
security of information exchange. Tiny sensor devices are contracts were also used to validate the service providers and
integrated with IoT for sensing, processing the information from service codes with minimum computational overhead on IoT
the dissimilar environmental conditions. Security is a devices. Finally, even throughput was achieved by utilizing a
challenging issue for authorized users able to access reliable

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

consortium blockchain and proof of authority. As one solution, for searching encryption to attest and monitor nodes [33]. The
Blockchain Structures (BSs) were designed to provide resource estimation was designed by a rule-based scaling
credibility verification [12]. BSswere is used for ensuring system in [34]. Unsupervised Deep Quadruplet Hashing with
response time, storage efficiency, and verification. However, Isometric Quantization (UDQH-IQ) was designed in [35] for
blockchains are found to be highly power-intensive, which solving the absence of local similarity of image pairs. In order
lowers throughput and limits power-constrained IoT devices. to improve the technology through IoT, crucial research studies
A blockchain system using a credit-based consensus mechanism and investigations have been done [36]. However, to achieve the
for IoT was presented in [13] to address these issues. This full potential of IoT lot of challenges and issues need to be
mechanism not only ensured system security but also provided addressed. The integration of blockchain technology with IoT
transaction efficiency concurrently. In order to overcome the was developed in [37]. However, intrinsic features of IoT result
issue, the novel blockchain model named BSMH-ZCD is in several challenges, namely decentralization, poor
developed to improve transaction efficiency. While a data interoperability, privacy, and security vulnerabilities.
authority management model was designed in [14] to safeguard Blockchain technology (BCT) is a promising solution [38] for
sensitive data confidentially, information security was providing security and protecting privacy on a large scale. A
achieved. A secure access control mechanism using the comprehensive survey of the existing blockchain protocols was
Ethereum blockchain for IoT networks was investigated in [15] introduced in [39] for the Internet of Things (IoT) networks. An
to measure temperature and intrusion detection. A more architecture for capability-based IoT access control utilizing the
thorough review of the use of blockchain with IoT is presented blockchain and decentralized identifiers is developed in [40] to
in [16]. A multi-agent simulation platform using certain manage the identity and access control for IoT devices.
management policies was designed in [17]. Performance However, the existing IoT access control solutions are mainly
patterns were analyzed via neighboring patterns to ensure based on conventional identity management schemes and use
security [18]. A taxi driver fraud detection system based on a centralized architectures. Blockchain and IoT as Blockchain of
routing mark is presented in [19]. An astringent analysis of taxi Things (BCoT) were analyzed in [41] for handling the issue of
mobility patterns with velocity values, spatiotemporal privacy, security, traceability, and reliability. Blockchain
distribution, hotspots, and trip durations proves to be of great Technology (BCT) -based applications were introduced in [42]
use for wireless researchers and government organizations [20]. for ensuring the security of the applications in IoT. Blockchain
A 3D clustering technique was utilized in [21] to obtain the technologies were investigated in [43] for IoT but, failed to
mobility patterns involved in the design of a taxi driver pattern improve the security. Deep Learning (DL) and blockchain-
system. The designed technique failed to reduce the running empowered security framework were introduced in [44] to
time. To address the issue, the BSMH-ZCD uses the Merkle enhance security and privacy. A blockchain-based decentralized
Hash Tree to minimize the running time. authentication modeling scheme called BlockAuth was
A method proposed in [22] utilizes IoT devices to obtain developed in [45] for achieving security performance. The
industrial-grade reliability for transferring information with the novel blockchain-based access control protocol named
aid of blockchain technologies. This method is a more efficient DBACP-IoTSG was introduced in [46] for evaluating the
transaction copy of the collected data from a smart home. The computational time. Blockchain-based security and privacy
method proposed in [23] utilizes a multivariate correlation systems were introduced in [48] to measure the performance of
analysis model to provide security for a smart home network. IoT. But, the running time was not considered. The
This model can analyze the traffic increase in the network and improvements of effective and efficient smart contract
identify the correlation between traffic features, therefore mechanisms were developed in [49] with higher security.
ensuring security and throughput to a greater extent. An However, it failed to balance the improved speed and security.
efficient deep CNN framework with hierarchical weighted The existing method of common issues such as failure to
fusion is introduced in [24] for surveillance videos. However, achieve higher security. However, computational complexity
the overhead was not reduced. A secure and fast image and running time, the packet delivery ratio was not decreased.
cryptosystem is presented in [25] for preserving the privacy of Many existing blockchain algorithms were developed to
patients but fails to enhance efficiency. A robust image hashing improve security. But, it failed to apply the Add Rotate XOR
with ring partition and invariant vector distance is introduced in block ciphers to enhance the security. The above state-of-the-
[26]. art works issue is addressed and motivated by, Merkle Hash
A novel trust assessment framework called STRAF was Zero Correlation Distinguisher for ARX block ciphers is
designed in [27] for estimating the trustworthiness of cloud presented in this paper. Then, the proposed model is needed for
services. A lightweight secure transport protocol denoted iTLS controlling personal information in smart city management
is introduced in [28] for end-to-end secure communication in services with higher security. The following sections provide a
IP-based IoT. Moreover, the Augmented Data Recognizing detailed description of the proposed method.
(ADR) algorithm is proposed with a wireless network [29], yet
III. PROBLEM DEFINITION
the processing time is not reduced. A new Chinese Remainder
Theorem (CRT)-based data storage mechanism was developed Many organizations lack the technical skills and expertise to
in [30] to store the user data securely in the cloud. However, the design a blockchain-based IoT system since Blockchain-as-a-
computational complexity was not reduced. To address the Service (BaaS) with IoT is still a developing area. Because
issue, the BSMH-ZCD applies the ZCD to minimize the blockchain-based IoT systems are only as secure as the system’s
computational complexity. access point, legal and compliance problems often occur in the
While an SDN-based decentralized security architecture was blockchain IoT networks. The issue of trust in blockchain-based
proposed to solve the security attack detection problems [31]. In IoT systems is difficult when no verification and audit
addition, an ARX block ciphers ZCD was designed in [32] to mechanism is utilized, particularly if they deal with sensitive
transform linear approximations into an integral distinguisher. information. In these systems, encryption of each object is
In a distributed network, a protocol is seamlessly incorporated required to prevent data leakages; thus, the issues in the

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

integration of IoT and blockchain need to be urgently addressed. citizens to reduce the overall criminal activities in the city.
Recently, much data has accumulated in blockchain-based IoT Smart contracts are effective small computer programs stored
systems, and only a limited part is valuable to obtain knowledge on a blockchain, which will perform a transaction under
and create action. Smart cities are deploying online services in specified conditions.
diverse sectors, such as transportation (intelligent road Instead of saving the IoT Novel Corona Virus patient data over
networks, connected cars, and public transport), public utilities the blockchain, a cloud storage server is used to save the traffic
(smart electricity, water, and gas distribution), education, health data. The first entity of cloud storage includes different
and social care, and public safety. Security from a hacker is the categorizes the users’ (taxi drivers and patients) data in
main concern in smart automation systems, whereby hackers indistinguishable blocks correlated with a distinctive block
can copy personal and official data if they gain access to a smart number, the cloud storage server. Figure 1 shows the sample
device. The other big concern with smart home systems is data blockchain framework.
privacy is due to the availability of consumer data by device and
platform manufacturers. This necessitates higher trust and
transparency between manufacturers and users of smart devices. 𝑏𝑙𝑜𝑐𝑘 1 𝑏𝑙𝑜𝑐𝑘 2 𝑏𝑙𝑜𝑐𝑘 𝑛

IV. SYSTEM SETUP Previous block Previous …. Previous

hash value block hash block hash
value value
Novel Corona Virus patient data is considered, The blockchain
system model includes ‘n’ number of transactions (i.e., ‘𝑇 = 𝑀𝑒𝑟𝑘𝑙𝑒 𝑡𝑟𝑒𝑒
𝑇1 , 𝑇2 , … . . , 𝑇𝑛 ’) and ‘n’ number of blocks (i.e., ‘𝐵 = 𝑀𝑒𝑟𝑘𝑙𝑒 𝑡𝑟𝑒𝑒
𝑀𝑒𝑟𝑘𝑙𝑒 𝑡𝑟𝑒𝑒
𝐵1 , 𝐵2 , … . . , 𝐵𝑛 ’ ). In each transaction, the hash value (i.e., left ….
𝑇𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑖𝑜𝑛𝑠 1
half ‘𝐻𝑟−1,𝐿 ’, right half 𝐻𝑟−1,𝑅 ) is created in the proposed 𝑇𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑖𝑜𝑛𝑠 2 𝑇𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑖𝑜𝑛𝑠 𝑛

BSMH-ZCD method by using the Merkle Hash tree. according

to the hash value, Zero Correlation Distinguisher is utilized to
perform data encryption, and decryption operations to obtain Fig.1 Sample blockchain framework
higher security in smart cities. Zero Correlation linear attack is
Figure 1 illustrates the sample blockchain framework which
a key improvement method for block ciphers. These algorithms
comprises the different blocks to form a chain. Each block of
are composed of simple addition, rotation, and XOR operations
the chain comprises a hash of the previous block hash value,
along with the ZCD that, in turn, support a robust
Merkle tree, and transaction. The blockchain has different
encryption/decryption mechanism for IoT devices. It is applied
transactions in the blockchain framework. Each transaction
to ensure robustness and, smooth transmission of blocks
comprises patient information collection from IoT devices. The
between IoT devices and the cloud server. Hence, the attack
hash values connect to blocks, which each has a single
complexity is reduced, and security is enhanced in smart cities.
transaction or thousands of transactions, even possessing
THREAT MODEL
thousands of hashes. To find one hash of the block, either an
Threat modeling is a significant part of software design. Threat
entire block can be combined, or a hash of each transaction can
modeling is a prearranged process to find security requirements,
be identified, which requires one hash. Then, the hash value is
and potential vulnerabilities, quantify threat and vulnerability
split into two portions, the left half, and the right half is created
criticality and prioritize remediation methods. In addition, the
by using the Merkle Hash tree. Once the data are stored in a
IoT risk is reduced. In IoT, the threat is the major important
block, the cloud server measures the hash of data in the single
issue. A threat is a potential negative action or event facilitated
block. According to the hash value, data encryption, and
by a vulnerability results.
decryption operation are carried out with the aid of the Zero
V. MERKLE HASH ZERO CORRELATION Correlation Distinguisher to improve the data security. Figure 2
DISTINGUISHER(ZCD) FOR ARXBLOCK CIPHERS shows the sample Merkle Tree with different hashes for eight
different transactions as follows.
With security as a continuing problem for IoT devices, the
predominantly uncontrolled IoT market leaves a great deal of In Figure 2, if one block contains eight transactions, then eight
room for device hacking, potentially compromising the secured hash values are identified, followed by the identification of the
and most sensitive data. When considering applications, such as hash values of one and two blocks, the hash values of three and
smart cities, this lack of security becomes increasingly serious. four blocks, and so on. Subsequently, the hash values of one,
For example, a road traffic hacker could take over a self-driving two, three, and four blocks are combined, and the hash value of
car with someone in it or may compromise data for fraudulent five, six, seven, and eight are combined. Finally, the hash values
activities based on the access levels given to an IoT system. of one, two, three, four, five, six, seven, and eight are combined,
With all of the data that is saved and transferred between IoT identifying one hash for an entire block. If the cloud server
devices, robust security is a prerequisite. Therefore, this work accepts the root hash of the new block, the new hash is appended
proposes a Blockchain Secured Merkle Hash Zero Correlation with the preceding hash value and produces the new hash of the
Distinguisher (BSMH-ZCD) for IoT in smart cities to address chain. This process eliminates the need for a third party.
the security aspect. The proposed BSMH-ZCD method includes
five different entities: cloud storage, peer-to-peer network,
smart city surveillance, smart contracts, and taxi drivers with
IoT devices.
The first entity in this proposed method is cloud storage, which
is used for storing data on multiple virtual servers that are
generally hosted by third parties. The city surveillance system
is one of the smart tools to ensure the safety and security of the

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

entire network. Each block size with a corresponding hash value

𝐻12345678
is split into two portions, the left half '𝐻𝑟−1,𝐿 ’, and the right half
‘𝐻𝑟−1,𝑅 ’. With these half portions, the addition ‘⊞’, rotation ‘>
𝐻1234 𝐻5678
>>’ and XOR ‘⨁’ operations are performed and added to the
ZCD. This is done to secure all the nodes so that not one single
node is compromised. The strength of ZCD is to support the
complex attack and, hence, is said to be highly secured.
𝐻12 𝐻34 𝐻56 𝐻78
Next, the pseudo-code representation of the ZCD for ARX
block data encryption is described below.

𝐻1 𝐻2 𝐻3 𝐻4 𝐻5 𝐻6 𝐻7 𝐻8 Algorithm1. Zero Correlation Distinguisher (ZCD) for ARX

Block Data Encryption
𝑇1 𝑇2 𝑇3 𝑇4 𝑇5 𝑇6 𝑇7 𝑇8
Input: Transaction‘𝑇 = 𝑇1 , 𝑇2 , … , 𝑇𝑛 ’, Block‘𝐵 =
Fig. 2 Eight-fold transaction Merkle Hash Tree 𝐵1 , 𝐵2 , … , 𝐵𝑛 ’
Output: Ciphertext‘𝐶’
The second entity in this proposed method is the peer-to-peer
network structure. Peer-to-peer (P2P) is a decentralized network Function ENCRYPTION (𝑑𝑎𝑡𝑎𝑓𝑖𝑙𝑒 )
model that has a number of devices (nodes) to store and share 1. If the user confirms data preservation over blockchain then
files, where every node acts as an individual peer(i.e., computer 2. Generate hash value for each transaction
system). P2P of blockchain permits the data to be transmitted 3. Let‘𝐶 ← 𝐸𝑛𝑐𝑟𝑦𝑝𝑡(𝑑𝑎𝑡𝑎𝑓𝑖𝑙𝑒 , 𝐻)’
worldwide without requiring intermediaries or a server. Then, 4. Else
the P2P network is shown in Figure 3. 5. Do nothing
6. End if
7. End function
𝑇3 𝑇2 𝑇1 𝐻12345678
𝐻𝑟−1,𝐿 𝐻𝑟−1,𝑅

𝐻5678
𝐻78 >>> 𝑟1

⊞
<<< 𝑟2

𝑇𝜆||𝜆′ (𝑟) ⨁
⨁
𝐻56
𝐻 𝐻𝑟,𝐿 𝐻𝑟,𝑅

Fig.4 Flow diagram of Zero Correlation Distinguisher (ZCD)

Fig.3 Peer-to-peer network frameworks for BSMH-ZCD for data encryption
In Figure 3, the p2p nodes are connected to the peer-to-peer This flow diagram 4 shows that the left half is represented by
network of either a computer, smartphone, or any other IoT ‘𝐻𝑟−1,𝐿 ’, and the right half is represented by ‘𝐻𝑟−1,𝑅 ’; the ZCD
device. Then, the blank box represents a computer, smartphone, key applied in the ‘𝑧𝑡ℎ’ round is represented by ‘𝑇𝜆||𝜆′ (𝑟)’.
or any other IoT device. In the BSMH-ZCD for IoT in smart
Then, the output of hash values for each corresponding two
cities, the peer-to-peer network is comprised of nodes and
halves is mathematically represented as:
proves that they are certified with a valid certificate before
making an account on the network via the cluster head. Once 𝐻𝑟,𝐿 = ((𝐻𝑟−1,𝐿 >>> 𝑟1 ) ⊞ 𝐻𝑟−1,𝑅 ) ⨁𝑇𝜆||𝜆′ (𝑟) (1)
authorized with the cluster head, the user signs data over the 𝐻𝑟,𝑅 = (𝐻𝑟−1,𝐿 <<< 𝑟2 )⨁𝐻𝑟,𝐿 (2)
network digitally. Each cluster has one cluster head that takes Next, to obtain the ZCD key assuming that ‘H’ is part of a
care of the nodes' key generation. The cluster head maintains cipher, a ZCD cipher ‘H’ is split the input into three portions
the keys of requesters (i.e., smart city surveillance) who can and the output into two parts. The three input parts and two
access the data of a specific taxi driver or patient and the public output parts are expressed as:
key of requesters (i.e., taxi drivers) that can be accessed. 𝐻′ = 𝐹2𝑎 ∗ 𝐹2 ∗ 𝐹2𝑏 → 𝐹2𝑐 ∗ 𝐹2𝑑 (3)
After obtaining the hash value for each data using Merkle Hash ′ (𝑝,
𝐻 𝑞, 𝑟) = (𝐻1′ (𝑝, 𝑞, 𝑟)𝐻2′ (𝑝, 𝑞, 𝑟)) (4)
Tree, valid certification is provided by applying symmetric key
encryption using the Zero Correlation Distinguisher (ZCD) for From (3) and (4), ‘𝑇𝜆||𝜆′ (𝑟): 𝐹2𝑏 → 𝐹2𝑐 ’ is defined as:
ARX block ciphers to encrypt the data for the blockchain. The 𝑇𝜆||𝜆′ (𝑟) = 𝐻1′ (𝜆, 𝜆′ , 𝑟) (5)
ZCD is a promising attack technique for block ciphers that are The function ‘𝑇𝜆||𝜆′ ’ is said to be the function ‘𝐻1′ ’ when the first
used to achieve efficient secure data access in IoT devices
through encryption and decryption operations. The proposed ‘𝑟’ bit and the ‘𝑟 + 1’ bit of its input are fixed to ‘𝜆’ and ‘𝜆′ ’,
method secures the node's network (i.e., individual IoT respectively. In this way, by splitting the target cipher of a
network) against attacks by intruders rather than securing the corresponding block hash value, the encryption and the

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

corresponding decryption process are said to be robust against The final entity of the proposed method comprises two types of
malicious nodes referred from [36]. datasets from taxi service drivers and Novel Corona Virus
The third entity is the smart city surveillance providers in the patients with wearable IoT devices. The IoT devices collect all
proposed method. it includes data on traffic or taxi drivers to data from the taxi driver, including trip ID, call type, origin of
perform surveillance or to offer data when malfunctioning has call, and origin stand from the drivers, and data. The proposed
been identified. Smart city surveillance providers deal with the ZCD (Zero Correlation Distinguisher) effectively transmits
analysis of fake activities once they receive an alert message information from sender to receiver with high security by using
from the network. Next, the integrated text is obtained using the the thread model and with minimum delay. The taxi drivers
pseudo-code shown below. themselves are the owners of their data and are responsible for
either granting or revoking data access from the smart city
Algorithm 2. Hash Block and Zero Correlation Sharing surveillance providers. Next, the pseudo-code representation of
Input: Transaction ‘𝑇 = 𝑇1 , 𝑇2 , … , 𝑇𝑛 ’, Block ‘𝐵 = Zero Correlation Distinguisher (ZCD) for block Data
𝐵1 , 𝐵2 , … , 𝐵𝑛 ’ Decryption is shown below.
Output: obtain integrated text
Function INTEGRATED_TEXT (𝑑𝑎𝑡𝑎𝑓𝑖𝑙𝑒 ) Algorithm 3. Zero Correlation Distinguisher (ZCD) for
1. If user choose anonymity over blockchain then Block Data Decryption
2. Generate a hash of the left and right portion Input: Encrypted file ‘𝐶’
3. Calculate ‘𝑇𝜆||𝜆′(𝑟) ’ Output: Decrypted ‘𝑑𝑎𝑡𝑎𝑓𝑖𝑙𝑒 ’
4. Mix hash of left and right portion with ZCD to form an Function DECRYPTION
integrated text 1. Let𝐺𝐶𝐷 ← 𝐷𝑒𝑐𝑟𝑦𝑝𝑡(𝐻𝑟−1,𝐿 , 𝐻𝑟−1,𝑅 )
5. End if 2. Let𝑑𝑎𝑡𝑎𝑓𝑖𝑙𝑒 ← 𝐷𝑒𝑐𝑟𝑦𝑝𝑡(𝐶, 𝑍𝐶𝐷)
6. End function 3. End function

This algorithm shows that the integration of a hash block and This algorithm shows that the main advantage of the ZCD for
zero-correlation sharing is used to form integrated text. If the blockchain technology is that it tracks billions of connected
user selects anonymity over the blockchain, then a hash of the devices and enables effective transaction processing and
left and right portions is either generated or does nothing. The coordination between devices. It has the additional advantage of
zero-correlation sharing is calculated upon the successful performing a distributed transaction ledger for several IoT
generation of a hash of the left and right portions. Finally, it is transactions at a time and is said to be computationally efficient.
integrated with the hashes of the left and right portions. This is why the distributed transaction ledger, being tamper-
The fourth entity is smart contracts in the proposed framework. proof, is said not to be manipulated by malicious users because
Smart contracts include generating agreements for any IoT it does not reside in a single location. This decentralized
devices, these are performed when given conditions are met. For approach, besides eradicating the single point of failure, makes
example, consider setting the conditions for a unique trip to the consumer or user data more private with the aid of
follow a certain route. Once an IoT device does not follow the cryptographic algorithms used by blockchains. Thus, IoT
indicated route, the smart contract will send an alert message to solutions ensure secure messaging between devices in an IoT
the authorized person or smart city surveillance provider. It will network by the BSMH-ZCD. Figure 6 demonstrates the flow
also store the abnormal route in the cloud so that the providers diagram of the proposed framework.
can also receive the taxi driver’s normal/abnormal route or a
patient’s normal/abnormal data if required for later use. Here,
Input Dataset
the actual decryption is performed. Figure 5 shows the flow
diagram of a Zero Correlation Distinguisher (ZCD) for data
decryption.
Initialize number of transaction and
block
Sender‘𝑆’ Receiver‘𝑅’

Generate hash value

Plaintext‘𝑃’inblocks Plaintext‘𝑃’in blocks

Apply Zero Correlation Distinguisher

Generate hashes For each block
for each block obtain hash value

Perform data encryption

𝐶 𝑃
= 𝐸𝑛𝑐𝑟𝑦𝑝𝑡 ൬𝐻𝑟,𝐿 = 𝐷𝑒𝑐𝑟𝑦𝑝𝑡 ൬𝐻𝑟,𝐿
Perform data decryption
∪ 𝐻𝑟,𝑅 (𝑇𝜆||𝜆′ (𝑟))൰ ∪ 𝐻𝑟,𝑅 (𝑇𝜆||𝜆′ (𝑟))൰

Achieve secure data access in IoT

Fig. 5 Flow diagram of Zero Correlation Distinguisher (ZCD) devices
for datad encryption
Figure 6 Flow chart of the proposed framework

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

Figure 6 shows the flow chart of the proposed framework for TABLE 2 EXPERIMENTAL SETUP
obtaining secure data access in IoT devices. Parameter Value
Number of Executions [2] 15,30,45,60,75,90, 105,120, 135,150
VI.EXPERIMENTSETUP Number of blocks [2] 9, 18, 27, 36, 45, 54, 63, 72, 81, 90
Number of taxi drivers [47] 442
In this work, taxi service trajectory data extracted from [47]
were used to conduct a fair performance analysis between the Number of IoT devices [47] 442
proposed BSMH-ZCD for IoT in smart cities and existing Number of Patient data[47] 100,200,300,400,500,600,700,800,900,1
000
methods, namely the TrustChain approach [1], SRB [2], and
DBACP-IoTSG [46]. The performance analysis was conducted
using the CloudSim simulator in the JAVA platform. The proposed BSMH-ZCD method takes a different number of
The purpose of the taxi service trajectory dataset is to increase executions (15 to 150) and blocks (9 to 90) from the dataset to
security, which consists of both the training and test data. It conduct the experimental evaluation. The experimental
includes an accurate dataset describing the complete year (from evaluation of the BSMH-ZCD method was compared against
01/07/2013 to 30/06/2014) of the (busy) trajectories performed three conventional works, namely the Trust Chain approach [1],
by 442 taxis running in Porto, Portugal. These taxis are said to SRB [2], and DBACP-IoTSG [46].
be operated with a taxi dispatch central device, using mobile VII.DISCUSSION
data terminals installed in the vehicles. The sizes of the training
and test data were 11296 and 437 KB, respectively. Comparative experiments were conducted to measure the
Each ride was classified into three classes: 1) taxi central based, performance of the proposed system using the introduced
2) stand-based, and 3) non-taxi central based. The first-class Blockchain Secured Merkle Hash Zero Correlation
provide an anonymized ID or information sets made available Distinguisher (BSMH-ZCD) for IoT in smart cities against three
from a telephone call. The second and third classes refer to existing methods, the TrustChain approach [1], SRB [2],
services that were demanded directly from the taxi drivers. Nine DBACP-IoTSG [46]. A performance analysis was conducted
features are included in each data sample that, in turn, for the following three metrics, Running time, Computational
correspond to one completed trip, as shown in Table 1. overhead, and Security
The experimental evaluation of BSMH-ZCD was conducted A. Performance analysis of running time
using a Novel Corona Virus 2019 Dataset reportedbyKaggle The time consumed to execute the overall process was taken as
[48]. The dataset includes eight files containing daily level data the first metric in the analysis of blockchain security for big data
on the number of affected cases, recovery, and deaths from generated by IoT applications in smart cities. In other words, the
Covid-19. It considers time-series records and the number of running time refers to the time consumed to generate the
cases on a given day as a collective number. We considered the security mechanism, which is mathematically formulated as:
COVID_open_line_list for the experiments. The dataset is 𝑅𝑇 = ∑𝑛𝑖=1 𝐸𝑥𝑒𝑖 ∗ 𝑇𝑖𝑚𝑒[𝐻𝑟,𝐿 + 𝐻𝑟,𝑅 ] (6)
comprised of 44 features, including ID, age, sex, city, country, where the running time ‘𝑅𝑇’ is recorded according to the
province, etc., and 13174 instances. number of executions ‘𝐸𝑥𝑒𝑖 ’ considered for conducting the
experiment and the time consumed for ensuring security. The
TABLE 1 TAXI SERVICE TRAJECTORY DATASET DATA SAMPLE time consumed is the summation of the hash generated for the
S.
Attributes Data type Description left ‘𝐻𝑟,𝐿 ’ and right half portions ‘𝐻𝑟,𝑅 ’, respectively. It is
No
1 Contains a unique identifier
measured in terms of milliseconds (ms).
TRIP_ID String
for each trip BSMH-ZCD [Proposed] Trust Chain approach [1]
2 Identifies the way used to
demand this service. 2.5 SRB [2] DBACP-IoTSG [46]
‘A’ – trip was dispatched from
central
CALL_TYPE Character 2
‘B’ – trip was demanded
Running time (ms)

directly to the taxi driver at a

specific stand 1.5
‘C’ – trip demanded randomly
3 It contains a unique identifier 1
ORIGIN_CALL Integer
for each phone number.
4 ORIGIN_STAN It contains a unique identifier
Integer 0.5
D for the taxi stand.
5 It contains a unique identifier
TAXI_ID Integer for the taxi driver that 0
performed each trip. 15 30 45 60 75 90 105 120 135 150
6 TIMESTAMP Integer Identifies the trip’s start. Number of executions
7 Identifies the day type of the
trip’s start.
‘B’ – trip started on a holiday
Fig.7 The running time under different numbers of executions
DAY TYPE Character or any other special day
‘C’ – trip started on a day Figure 7 shows the performance results of the running time
before type ‘B’ under different numbers of executions performed by different
‘A’ – normal or workday numbers of taxis running in the city of Porto.
8 False – when GPS data stream
MISSING_DAT is complete
Boolean
A True – when one or more
locations are missing
9 POLYLINE String List of GPS coordinates

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

memory consumed while analyzing the security aspect. The

BSMH-ZCD [Proposed] Trust Chain approach [1]
memory consumed involves the summation of the hash
40
SRB [2] DBACP-IoTSG [46] generated for the left ‘𝐻𝑟,𝐿 ’ and right half portions ‘𝐻𝑟,𝑅 ’,
35 respectively. It is measured in kilobytes (KB).
Running Time (ms)

30
25
20 BSMH-ZCD [Proposed] Trust Chain approach [1]
15
350 SRB [2] DBACP-IoTSG [46]
10

Computational overhead (KB)

5 300
0 250
100 200 300 400 500 600 700 800 900 1000 200
No of executions
150

Fig.8. The running time under different numbers of executions 100

50
Figure 8 displays the running time with a number of patient data
0
in the range of 100 to 1000 for Novel Corona Virus 2019 15 30 45 60 75 90 105 120 135 150
Dataset. The lower the running time, the higher the number of Number of executions
data files waiting in the cloud server, according to which the
blocks are said to be performed in a significant manner.
Fig.9. The computational complexity of different executions
However, with a higher running time, only a limited number of
blocks are said to be executed and stored in the cloud server. Figure 9 illustrates the computational complexity involved in
Comparatively, with a higher number of executions, a larger generating the encryption and decryption mechanisms. Figure
number of IDs must be processed; hence, the higher running 10 shows the computational overhead results with an increasing
time is noted. The preceding graphical representation, refer to number of patient data using the Novel Corona Virus 2019
Fig. (8) based on the mathematical calculations performed, Dataset. However, it is observed from the sample calculations
shows improved results attained using the proposed BSMH- that the computational overhead is comparatively less when
ZCD. In the above figure 7, let us consider ‘9’ number of applied with BSMH-ZCD than with the [1] [2] and [46]. In the
executions for conducting the experiments, the time consumed above figure 9, ‘9’ patient data is considered to evaluate the
by the algorithm to improve the security is 1.1025𝑚𝑠’, whereas experiments, the memory consumed by the algorithm to
‘1.1125𝑚𝑠, ‘‘1.4025𝑚𝑠,’ and ’ 1.5025𝑚𝑠 of time consumed enhance the the security is ‘45kb’, whereas ‘60𝑘𝑏, ‘’75 𝑘𝑏,’
by existing techniques [1], [2] and [46]. As revealed in the chart, and ’ 90 𝑘𝑏 of time consumed by existing techniques [1], [2]
the running time is gradually increased for all three blockchain and [46]. As revealed in the chart, the computational overhead
methods while raising the number of patient data since the is gradually increased while increasing the number of patient
counts of data get increased for each run. This is contributed to data for each run.
the usage of the blockchain for IoT devices that provides
security and also computes the computationally efficient
encryption mechanism using the Merkle Hash Tree.This Merkle BSMH-ZCD [Proposed] Trust Chain approach [1]
Hash Tree structure permits effective mapping of big data, SRB [2] DBACP-IoTSG [46]
100
making it easier to identify minor updating of the prepared data.
Computational Overhead (kb)

If the occurrence of data change must be observed, such as if 80

the taxi driver's route changes, this is done in the root hash,
therefore avoiding the traversal of the entire tree. In other words, 60
data consistency is measured by only traversing a small part of
the structure instead of the entire tree with the aid of hashing, 40
which minimizes the running time of the encryption process.
20
Herein, using the BSMH-ZCD method for the taxi driver data,
the running time was reduced by 15%, 42% and 52% compared 0
to [1], [2] and [46] respectively. Similarly, for the Novel Corona 100 200 300 400 500 600 700 800 900 1000
Virus 2019 Dataset, the BSMH-ZCD method reduced running
No of Patient Data
time by 20%, 32% and 38% compared to the [1], [2] and [46]
respectively.
B. Performance analysis of computational overhead Fig.10 The computational complexity under different numbers
The second metric used in the blockchain security analysis for of patient data
big data generated by IoT applications in smart cities is the This is an attribute to the application of the Merkle Hash Tree
memory consumed in executing the overall process of security in the blockchain component and the Zero Correlation
analysis. In other words, the computational overhead refers to Distinguisher (ZCD). Specifically, the Merkle Hash Tree
memory consumed while evaluating the security mechanism. creates the hash value for each transaction, then the output hash
This is mathematically formulated as: sum length is the same as that measured by the hashing
𝐶𝑂 = ∑𝑛𝑖=1 𝐸𝑥𝑒𝑖 ∗ 𝑀𝐸𝑀𝑂𝑅𝑌[𝐻𝑟,𝐿 + 𝐻𝑟,𝑅 ] (7) algorithm. Therefore, large amounts of data (i.e., big data) are
From equation (7), the computational overhead ‘𝐶𝑂’ is obtained said to be identified (i.e., encrypted/decrypted) exclusively by
based on the executions being conducted ‘𝐸𝑥𝑒𝑖 ’ and the

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

utilizing their resulting hash. After obtaining the hash for each
BSMH-ZCD [Proposed] Trust Chain approach [1]
data through the Merkle Hash Tree, the Zero Correlation 90
Distinguisher performs the data encryption and decryption SRB [2] DBACP-IoTSG [46]
80
operation for the blockchain based on the hash value. Thus, with

Packet Delivery Ratio (%)

the involvement of big data in the proposed method, the 70
advantages of being able to store and identify data specifically 60
for encryption/decryption with fixed-length output results in 50
extensive storage savings (i.e., computational overhead) and 40
helps to increase efficiency. As a result, the computational 30
overhead incurred using BSMH-ZCD was found to be reduced 20
by 11%, 21% and 38% compared to [1], [2] and [46]. By
10
applying the Novel Corona Virus 2019 Dataset, BSMH-ZCD
reduced computational overhead by 24%, 28% and 32% 0
100 200 300 400 500 600 700 800 900 1000
compared to [1], [2] and [46] respectively.
No of Patient Data
C. Performanceanalysisofsecurity
Finally, the security is measured in terms of packet delivery Fig.12 The packet delivery ratio under different numbers of
ratio ‘𝑃𝐷𝑅’. Packet delivery ratio in the work is the percentage patient data
ratio of the number of data packets (i.e., blocks) successfully This is because the incorporation of the ZCD for ARX block
received by the cloud server ‘𝐷𝑟𝑒𝑐 ’ to the total number of data data encryption/decryption can ensure a double layer of
packets sent ‘𝐷𝑡𝑜𝑡𝑎𝑙 ’ by the source IoT device. This is expressed security. Specifically, the ZCD in the proposed BSMH-ZCD
as: model is integrated by the hashes of the left and right portions.
𝐷
𝑃𝐷𝑅 = 𝑟𝑒𝑐 ∗ 100 (8) The proposed method secures the individual IoT network
𝐷𝑡𝑜𝑡𝑎𝑙
A higher packet delivery ratio ensures that a higher number of against attacks by intruders rather than securing the entire
blocks are being sent to the cloud server; hence, the higher network. Next, the ZCD is applied to ensure robustness and, in
packet or block is said to be secure. Conversely, a lower packet turn, the smooth transmission of blocks between the source IoT
delivery ratio indicates that a smaller number of blocks are device and the cloud server. In addition to eliminating the
being sent to the cloud server, and thus, the lower packet or requirement of a third party via blockchain, the packet delivery
block is said to be secure. ratio is said to be increased. ZCD can effectively reduce the
attack complexity, therefore increasing the security in smart
cities. Compared to [1], [2] and [46], the BSMH-ZCD model
BSMH-ZCD [Proposed] Trust Chain approach [1]
improved security by 9%, 23% and 35% and increased the
100 SRB [2] DBACP-IoTSG [46] packet delivery ratio by 15%, 23% and 37% by using Novel
Packet delivery ratio (%)

80
Corona Virus 2019 Dataset, respectively.
VIII CONCLUSION
60
This work introduces the Blockchain Secured Merkle Hash Zero
40 Correlation Distinguisher (BSMH-ZCD) model as a solution to
security threats while considering the computational limitations
20
of IoT. The proposed model was designed with the innovation
0 of Merkle Hash Tree and Zero Correlation Distinguisher. For
9 18 27 36 45 54 63 72 81 90 each transaction, the Merkle Hash Tree produces a hash value,
Number of blocks then the Zero Correlation Distinguisher performs the data
encryption and decryption operation with the left and right
Fig.11 Security analyses via packet delivery ratio portions of the hash value. It is deal with personal information
in smart city management services with minimum run time. The
Figure 11 shows the performance graphical analysis of security, ARX block cipher or the addition, rotation, and XOR operations
and Figure 12 displays the packet delivery ratio with different are performed to achieve efficient secure data access in the IoT
numbers of patient data for the Novel Corona Virus 2019 devices via encryption and decryption. Therefore, the attack
Dataset. The higher the packet delivery ratio, the higher the complexity is minimized in smart cities. The BSMH-ZCD
security is vice versa. According to the mathematical model provides better performance of improved security and
calculations and graphical representation, the packet delivery reduced the running time and computational overhead.
ratio was found to be higher by applying BSMH-ZCD when However, because the privacy preservation rate is not
compared to [1] and [2]. Let us consider 9 data blocks in the considered in the discussion section, our future works will
first iteration. By applying the BSMH-ZCD model 8 of blocks examine blockchain-based decentralized trust management and
are being sent to the cloud server then the packet delivery ratio secure the usage control scheme of IoT big data to solve the trust
is 88.88%. The 77.77%, 66.66% and 55.55% are achieved by and security issues of IoT big data management. The designed
using existing [1], [2] and [46]. Likewise, ten runs are method comprises data gathering, invoking, transfer, storage,
performed and the results are observed for each method. The and usage processed over the blockchain smart contract. A huge
performance of the proposed BSMH-ZCD model is compared amount of data is used and the performance of BlockBDM
to other existing methods. scheme is providing feasible, secure and scalable for
decentralized trust management of IoT big data. It is used to
improve the storage trust.

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

Appendix. Audit and Provenance in Mobile Cloud." Mathematical Problems in

Engineering (2019).
The proposed model symbols and description is mentioned in
[15] Ghadekar, Premanand, NiketDoke, SushmitaKaneri, and VarshaJha.
table A.1. “Secure Access Control to IoT Devices Using Blockchain.” International
Journal of Recent Technology and Engineering (IJRTE) 8, no. 2 (July
TABLE A.1. SYMBOLS AND DESCRIPTION 2019): 1–7.
[16] Fernández-Caramés, Tiago M., and Paula Fraga-Lamas. "A Review on the
Symbol Description
Use of Blockchain for the Internet of Things." IEEE Access 6 (2018):
‘𝑇 = 𝑇1 , 𝑇2 , … . . , 𝑇𝑛 ’ Number of Transactions 32979-33001.
‘𝐵 = Number of Blocks [17] Cheng, Shih-Fen, and Thi Duong Nguyen. "Taxisim: A multiagent
𝐵1 , 𝐵2 , … . . , 𝐵𝑛 ’ simulation platform for evaluating taxi fleet operations." In Proceedings of
‘𝐻𝑟−1,𝐿 ’ Left Half the 2011 IEEE/WIC/ACM International Conferences on Web Intelligence
and Intelligent Agent Technology-Volume 02, pp. 14-21. IEEE Computer
‘𝐻𝑟−1,𝑅 ’ Right Half Society, 2011.
‘𝑇𝜆||𝜆′ (𝑟)’ ‘𝑧𝑡ℎ’ round [18] Deng, Zhongwei, and Minhe Ji. "Spatiotemporal structure of taxi services
in Shanghai: Using exploratory spatial data analysis." In 2011 19th
‘𝑟’ bit International Conference on Geoinformatics, pp. 1-5. IEEE, 2011.
‘𝑟 + 1’ bit of its input are fixed to ‘𝜆’ and ‘𝜆′’ [19] Ge, Yong, Hui Xiong, Chuanren Liu, and Zhi-Hua Zhou. "A taxi driving
‘⊞’ Addition fraud detection system." In 2011 IEEE 11th International Conference on
Data Mining, pp. 181-190. IEEE, 2011.
‘>>>’ Rotation [20] Hoque, Mohammad Asadul, Xiaoyan Hong, and Brandon Dixon. "Analysis
‘⨁’ XoR of mobility patterns for urban taxi cabs." In 2012 international conference
‘𝐶’ Ciphertext on computing, networking and communications (ICNC), pp. 756-760.
IEEE, 2012.
‘𝑅𝑇’ Running Time [21] Liu, Liang, Clio Andris, AssafBiderman, and Carlo Ratti. "Revealing taxi
‘𝐸𝑥𝑒𝑖 ’ Number of Executions driver's mobility intelligence through his trace." In Movement-Aware
‘𝐶𝑂’ Computational Overhead Applications for Sustainable Mobility: Technologies and Approaches, pp.
105-120. IGI Global, 2010.
‘𝑃𝐷𝑅’ Packet Delivery Ratio
[22] Skwarek, Volker. "Blockchains as security-enabler for industrial IoT-
‘𝐷𝑟𝑒𝑐 ’ Number of data packets (i.e., blocks) applications." Asia Pacific Journal of Innovation and Entrepreneurship 11,
successfully received by the cloud server no. 3 (2017): 301-311.
[23] Singh, Saurabh, In-Ho Ra, WeizhiMeng, Maninder Kaur, and Gi Hwan
‘𝐷𝑡𝑜𝑡𝑎𝑙 ’ Total number of data packets sent
Cho. "SH-BlockCC: A secure and efficient Internet of things smart home
architecture based on cloud computing and blockchain technology."
REFERENCES International Journal of Distributed Sensor Networks 15, no. 4 (2019):
[1] Jayasinghe, Upul, GyuMyoung Lee, ÁineMacDermott, and Woo Seop 1550147719844159.
Rhee. "TrustChain: A Privacy Preserving Blockchain with Edge [24] Khan Muhammad, Tanveer Hussain, Mohammad Tanveer, Giovanna
Computing." Wireless Communications and Mobile Computing (2019). Sannino, Victor Hugo C., “Cost-Effective Video Summarization Using
[2] Huang, Ke, Xiaosong Zhang, Yi Mu, FatemehRezaeibagha, Xiaojiang Du, Deep CNN With Hierarchical Weighted Fusion for IoT Surveillance
and NadraGuizani. "Achieving Intelligent Trust-Layer for IoT via Self- Networks”, IEEE Internet of Things Journal, v. 7, pp. 4455-4463, 2020.
Redactable Blockchain." IEEE Transactions on Industrial Informatics [25] Rafik Hamza, Zheng Yan, Khan Muhammad, Paolo Bellavista,
(2019). FaizaTitouna, “A privacy-preserving cryptosystem for IoT E-healthcare”,
[3] Li, Shuling. "Application of blockchain technology in smart city Information Sciences, pp-1-18.
infrastructure." In 2018 IEEE International Conference on Smart Internet [26] Zhenjun Tang, Xianquan Zhang, Xianxian Li, and Shichao Zhang, “Robust
of Things (SmartIoT), pp. 276-2766. IEEE, 2018. Image Hashing with Ring Partition and Invariant Vector Distance”, IEEE
[4] Ramezan, Gholamreza, and Cyril Leung. "A Blockchain-Based Contractual Transactions on information forensics and security, v. 11, no. 1, 2016.
Routing Protocol for the Internet of Things Using Smart Contracts." [27] Xiang Li, Qixu Wang, Xiao Lan, Xingshu Chen, Ning Zhang, Dajiang
Wireless Communications and Mobile Computing (2018). Chen, “Enhancing Cloud-Based IoT Security through Trustworthy Cloud
[5] Ishmaev, Georgy. "The Ethical Limits of Blockchain-Enabled Markets for Service: An Integration of Security and Reputation Approach”, IEEE
Private IoT Data." Philosophy & Technology (2019): 1-22. Access, v.7, pp 9368-9383,2019.
[6] Sun, Jianjun, Jiaqi Yan, and Kem ZK Zhang. "Blockchain-based sharing [28] Pengkun Li, Jinhu Su and Xiaofeng Wang, “iTLS: Lightweight Transport
services: What blockchain technology can contribute to smart cities." Layer Security Protocol for IoT with Minimal Latency and Perfect Forward
Financial Innovation 2, no. 1 (2016): 26. Secrecy”, IEEE Internet of Things Journal, v.7, no.8, pp 6828-6841, 2020.
[7] Kortesniemi, Yki, DmitrijLagutin, TommiElo, and Nikos Fotiou. [29] RajmohanPalanivelu, Srinivasan P.S.S., “Safety and security measurement
"Improving the Privacy of IoT with Decentralised Identifiers (DIDs)." in industrial environment based on smart IoT technology based augmented
Journal of Computer Networks and Communications 2019 (2019). data recognizing scheme”, Computer Communications, v.150, pp-777-
[8] Biswas, Sujit, Kashif Sharif, Fan Li, BoubakrNour, and Yu Wang. "A 787, 2020.
Scalable Blockchain Framework for Secure Transactions in IoT." IEEE [30] BalasubramanianPrabhukavin, SannasiGanapathy, “A secured storage and
Internet of Things Journal (2018). privacy-preserving model using CRT for providing security on cloud and
[9] Mehedi, SK Tanzir, Abdullah Al MamunShamim, and Mohammad IoT-based applications”, Computer Networks, v.151, pp-777-787, 2019.
BadrulAlam Miah. "Blockchain-based security management of IoT [31] Shailendra Rathore, ByungWook Kwon, Jong Hyuk Park,
infrastructure with Ethereum transactions." Iran Journal of Computer “BlockSecIoTNet: Blockchain-based decentralized security architecture
Science 2, no. 3 (2019): 189-195. for IoT network”, Journal of Network and Computer Applications, v.143,
[10] Jesus, Emanuel Ferreira, Vanessa RL Chicarino, Célio VN de Albuquerque, pp-167-177, 2019.
and Antônio A. de A. Rocha. "A survey of how to use blockchain to secure [32] Long Wen, Meiqin Wang. “Integral Zero-Correlation Distinguisher for
internet of things and the stalker attack." Security and Communication ARX Block Cipher, with Application to SHACAL-2”, Information Security
Networks (2018). and Privacy, Springer, 2014, pp- 454-461.
[11] Xu, Yang, Guojun Wang, Jidian Yang, Ju Ren, Yaoxue Zhang, and Cheng [33] Cai, Chengjun, Xingliang Yuan, and Cong Wang. "Hardening distributed
Zhang. "Towards secure network computing services for lightweight clients and encrypted keyword search via blockchain." In 2017 IEEE Symposium
using blockchain." Wireless Communications and Mobile Computing on Privacy-Aware Computing (PAC), pp. 119-128. IEEE, 2017.
(2018). [34] Q. Z. Ullah, G. M. Khan and S. Hassan, "Cloud Infrastructure Estimation
[12] Qu, Chao, Ming Tao, Jie Zhang, Xiaoyu Hong, and Ruifen Yuan. and Auto-Scaling Using Recurrent Cartesian Genetic Programming-Based
"Blockchain based credibility verification method for IoT entities." ANN," in IEEE Access, vol. 8, pp. 17965-17985, 2020,
Security and Communication Networks (2018). [35] Qibing Qin, Lei Huang, Zhiqiang Wei, JieNie, KezhenXie, JinkuiHou,
[13] Huang, Junqin, Linghe Kong, Guihai Chen, Min-You Wu, Xue Liu, and Unsupervised Deep Quadruplet Hashing with Isometric Quantization for
Peng Zeng. "Towards secure industrial IoT: Blockchain system with credit- image retrieval, Information Sciences, Volume 567,2021, Pages 116-130,
based consensus mechanism." IEEE Transactions on Industrial Informatics ISSN 0020-0255,
(2019). [36] Kumar, S., Tiwari, P. &Zymbler, M. Internet of Things is a revolutionary
[14] Zheng, Rongyue, Jianlin Jiang, XiaohanHao, Wei Ren, Feng Xiong, and Yi approach for future technology enhancement: a review. J Big Data 6, 111
Ren. "bcBIM: A Blockchain-Based Big Data Model for BIM Modification (2019).

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JIOT.2022.3171237, IEEE
Internet of Things Journal

IEEEINTERNETOFTHINGS,VOL.XX,NO.XX,MM2020

[37] H. -N. Dai, Z. Zheng and Y. Zhang, "Blockchain for Internet of Things: A
Survey," in IEEE Internet of Things Journal, vol. 6, no. 5, pp. 8076-8094,
Oct. 2019,
[38] W. Viriyasitavat, L. D. Xu, Z. Bi and D. Hoonsopon, "Blockchain
Technology for Applications in Internet of Things—Mapping From System
Design Perspective," in IEEE Internet of Things Journal, vol. 6, no. 5, pp.
8155-8168, Oct. 2019
[39] M. A. Ferrag, M. Derdour, M. Mukherjee, A. Derhab, L. Maglaras and H.
Janicke, "Blockchain Technologies for the Internet of Things: Research
Issues and Challenges," in IEEE Internet of Things Journal, vol. 6, no. 2,
pp. 2188-2204, April 2019
[40] Yue Liu, Qinghua Lu, Shiping Chen, Qiang Qu, Hugo O’Connor, Kim-
Kwang Raymond Choo, He Zhang, Capability-based IoT access control
using blockchain, Digital Communications and Networks,2020, ISSN
2352-8648
[41] Hong-Ning Dai, Zibin Zheng, Yan Zhang, “Blockchain for Internet of
Things: A survey”, IEEE Internet of Things Journal, v. 6, no.5, pp-8076 –
8094, 2019
[42] Wattana Viriyasitavat, Li Da Xu; Zhuming Bi, Danupol Hoonsopon,
“Blockchain Technology for Applications in Internet of Things—Mapping
From System Design Perspective”, IEEE Internet of Things Journal,v.6,
no. 5, pp- 8155 – 8168, 2019
[43] Mohamed Amine Ferrag, Makhlouf Derdour, Mithun Mukherjee,
Abdelouahid Derhab, Leandros Maglaras, Helge Janicke, “Blockchain
technologies for the internet of things: Research issues and challenges”,
IEEE Internet of Things Journal, v.6, no.2, pp- 2188 – 2204, 2019
[44] Shailendra Rathore, Jong Hyuk Park, Hangbae Chang, “Deep Learning and
Blockchain-Empowered Security Framework for Intelligent 5G-Enabled
IoT”, IEEE Access, v.9, pp- 90075 – 90083,2021.
[45] Ma Zhaofeng, Meng Jialin, Wang Jihui, Shan Zhiguang, “Blockchain-
Based Decentralized Authentication Modeling Scheme in Edge and IoT
Environment”, IEEE Internet of Things Journal, v. 8, no. 4, pp- 2116 –
2123, 2021
[46] Basudeb Bera, Sourav Saha, Ashok Kumar Das, Athanasios V. Vasilakos,
“Designing Blockchain-Based Access Control Protocol in IoT-Enabled
Smart-Grid System”, IEEE Internet of Things Journal, v. 8, no.7, pp- 5744
– 5761, 2021.
[47] Luis Moreira-Matias "Taxi Service Trajectory Dataset" URL:
http://www.geolink.pt/ecmlpkdd2015challenge/dataset.html (Last
accessed on 02 Feb 2020)
[48] Kaggle.com. 2020. Novel Corona Virus 2019 Dataset. [online] Available
at: <https://www.kaggle.com/sudalairajkumar/novel-corona-virus-2019-
dataset> [Accessed 20 September 2020].
[49] Mohamed Amine Ferrag, Lei Shu, “The Performance Evaluation of
Blockchain-based Security and Privacy Systems for the Internet of Things:
A Tutorial”, IEEE Internet of Things Journal, 2021, pp-1-25.
[50] Security Challenges and Opportunities for Smart Contracts in Internet of
Things: A Survey”, IEEE Internet of Things Journal, v. 8, no. 15, 2021, pp-
12004 – 12020.

2327-4662 (c) 2021 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: Technische Universitaet Muenchen. Downloaded on May 28,2022 at 14:12:23 UTC from IEEE Xplore. Restrictions apply.