Received 12 December 2020; accepted 11 January 2021.

Date of publication 20 January 2021;

date of current version 12 February 2021. The review of this article was arranged by Associate Editor Peng Li.
Digital Object Identifier 10.1109/OJCS.2021.3053032

A Survey on Privacy-Preserving Blockchain

Systems (PPBS) and a Novel PPBS-Based
Framework for Smart Agriculture
QUANG NHAT TRAN 1 , BENJAMIN P. TURNBULL 1 , HAO-TIAN WU 2 , A. J. S. DE SILVA3 ,
KATERINA KORMUSHEVA3 , AND JIANKUN HU 1 (Senior Member, IEEE)
1
UNSW Canberra, The University of New South Wales, Canberra, ACT 2600, Australia
2
School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China
3
Telsoft Limited, Canberra 2586, Australia
CORRESPONDING AUTHOR: JIANKUN HU (e-mail: [email protected])
This work was supported by ARC funds under Grants DP190103660, DP200103207, and LP180100663.

ABSTRACT Blockchain and smart contracts have seen significant application over the last decade, revolu-
tionising many industries, including cryptocurrency, finance and banking, and supply chain management.
In many cases, however, the transparency provided potentially comes at the cost of privacy. Blockchain
does have potential uses to increase privacy-preservation. This paper outlines the current state of privacy
preservation utilising Blockchain and Smart Contracts, as applied to a number of fields and problem domains.
It provides a background of blockchain, outlines the challenges in blockchain as they relate to privacy, and
then classifies into areas in which this paradigm can be applied to increase or protect privacy. These areas
are cryptocurrency, data management and storage, e-voting, the Internet of Things, and smart agriculture.
This work then proposes PPSAF, a new privacy-preserving framework designed explicitly for the issues that
are present in smart agriculture. Finally, this work outlines future directions of research in areas combining
future technologies, privacy-preservation and blockchain.

INDEX TERMS Biometrics, blockchain, consensus protocol, internet of things, privacy, privacy-
preservation, smart agriculture, smart energy.

I. INTRODUCTION a transformative new platform like blockchain, which enables

Far from its origins in cryptocurrency, blockchain has shown enhanced information, and guarantees the quality of informa-
its applicability in various fields, including supply chain man- tion, without the endogeneity effect of advertising, enables
agement [1]–[6], biomedical healthcare [7]–[10], cloud com- a very different role for marketing. However, the nature of
puting [11], [12], identity management [13], [14], market- blockchain does not immediately consider user or data pri-
ing [15]–[17], and even tourism [18]. An interesting different vacy, and the paradigm structure poses a challenge for these.
field of application is in marketing, where the authors of [19], Sensitive information leakage may have significant impacts
illustrate how blockchain technology can act to empower the on the use-case of the system; this may include information
consumer-centric paradigm. By fostering disintermediation, leakage, data exposure or the exposure of party identities.
blockchain can act as the mechanism to create traceability For instance, a blockchain-enabled vehicle infrastructure plat-
and transparency. This also links to the latest evidence ques- form would, by its very nature, expose locations, transactions
tioning the effectiveness of advertising, and in particular dig- and other Personally Identifiable Information (PII), for which
ital advertising [19], [20]. Repeated studies have shown TV there are potentially serious impacts. However, there has been
ads [21] and search words digital advertising [20] to be not as research in the intersection of blockchain and multiple forms
effective as suggested by the advertisement empires that run of privacy preservation; both general, and applied to spe-
them. The future of linking producers and consumers through cific areas. Hence, there is a necessity to collate, review and

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
72 VOLUME 2, 2021
categorize these published works on privacy preservation as the predetermined amount of cryptocurrency to be locked up.
applied to different Blockchain applications. When chosen, the user is required to propose a block, which is
In this survey, we first outline the general structure of validated by the validators. If this block is valid, they receive
a Blockchain and some of its current privacy issues in a reward. If the user is found dishonest, the stake is taken. The
Section II. Afterward, Section III of this work categorizes system works as, given an input, any node in the network can
the studies that have been proposed in privacy-preserving verify the validity of the block to update the blockchain.
blockchain, and reviews the key advances. Based on the On the other hand, the ledger is used for the main purpose
outcomes, a categorization on the privacy preservation in of blockchain: deliver data between parties. Each party’s ad-
Blockchain is established and summarized. Then, a novel dress is a unique pseudonym in the network, generated by
PPBS-based framework is proposed for smart framing. Fi- the PKI, ensuring the identity of the involved parties. When a
nally, Section VI provides a summary of predictions on the communication happens, a transaction is created to record the
future trends of privacy preservation as applied to Blockchain addresses, the signatures of the parties, and the data. A block
technologies. is generated, carrying the hash value of the previous block
and a timestamp, to store transaction data in a Merkle tree.
II. BLOCKCHAIN TECHNOLOGY Note that the first block in the chain does not have previous
Blockchain was first proposed by Chaum [22] in 1979. In block. Due to this structure, blockchain data is robust against
1992, Bayer et al. [23] improved this concept by incorporating tampering attacks.
Merkle trees as part of the block design. The structure is The term ‘Smart contract’ is a multi-definition term that
an ever expanding list of blocks that are cryptographically is dependent on how it is used and implemented in each
connected: each block contains the previous block’s crypto- framework. Generally, a smart contract provides mechanisms
graphic hash, a timestamp, and the main data, which is also re- for documenting and executing contracts without the need for
ferred to by transaction data. Owing to this design, blockchain a human trusted intermediary. There are several well-known
is strongly resistant against data modification. Year of 2008 smart contract platforms, including Ethereum, Hyperledger
marked the day when blockchain commenced its technology Fabric, Nem, and Stellar, all of which are based on, but extend,
widespreading by being conceptualized in Nakamoto’s stud- the Blockchain paradigm.
ies. In all of its applications, the purpose of blockchain is to There are three distinct forms of blockchain, to account for
deliver data to parties with integrity. different use-cases. These are:
r Public: A public blockchain grants access to all partic-
A. HOW BLOCKCHAIN WORKS ipants as they can read the data from it or even create
Blockchain is an immutable database in which only new a new transaction to attach to it. Hence, this type of
records (or blocks) can be appended. A peer-to-peer network blockchain is accessible even to an adversary. Public
manages the communication for the chain of blocks, allowing blockchains are the most common and include cryp-
applications to access the data contained in each block by tocurrencies, particularly Bitcoin, Dogecoin as well as
broadcasting the whole structure to all the nodes based on a smart contract platforms such as Ethereum.
consensus algorithm. As such, a blockchain does not require r Consortium: A consortium (or semi-private) blockchain
a central server, making decentralization one of its advantages grants a number of parties the right to confirm blocks.
for deployment. In order to add a new block to the chain, a In a consortium blockchain, the consensus algorithm is
consensus algorithm is used such that all nodes in the net- operated by a number of nodes in the network. Partici-
work (for a public blockchain) or the managing entities (for a pants may still read the data, but not the general public.
consortium/private blockchain) must validate the transactions Consortium blockchains are used for cases when there
before creating the block. This prevents the issue of double- are several related, but not mutually trusted, entities that
spending, which is the risk that a unit of digital currency is are working together or require data sharing.
spent twice due to digital information replication. r Private: A private blockchain retains the write privilege
There are two most famous consensus algorithms used in to an entity or organization while the read permission
blockchain: Proof of Work (PoW) and Proof of Stake (PoS). might be open to public or to certain parties. Private
Conceptually, the miners compete against one another to look blockchains are often applied in government or banking
for the solution of a mathematically complicated problem. disciplines where access control must be monitored, but
In PoW, the miners use their own computational power to where other data storage processes are less optimal.
calculate the hash value of the blockchain data and a vari- Each of these forms of blockchain has its own advantages or
able data (called nonce in Bitcoin). This hash value must disadvantages. While public blockchains are entirely decen-
satisfy certain conditions given by the protocol. As a miner tralized, the information contained is exposed to adversaries.
finds the valid hash, he/she gets rewarded with cryptocurrency Consortium blockchains are effective when data sharing is
for the computational power spent. Due to low efficiency of needed. However, its security depends on the architecture
computational power, PoS was proposed. In PoS, there is and security techniques outside of the paradigm. Last but not
no mining, thus, no miners. Users in the system are chosen least, private blockchains remove the decentralization char-
based on some factors. One of these is called “stake,” which is acteristics. As a result, a casual centralized data storage and

VOLUME 2, 2021 73
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

TABLE 1. Privacy Categorization

processing systems tend to have more benefits over the use The need for, and types of privacy, vary per application.
of a private blockchain. There are some advantages to private Based on its nature, the privacy is classified into: Biometric
blockchain implementations; for example, private blockchains Privacy, Identity Privacy (consisting of Personal and Object
have the benefit of transaction transparency, for example, Privacy), and Data Privacy. The relation between the types of
which can provide numerous benefits for specific use-cases. application and privacy is presented in Table 1.

B. CURRENT PRIVACY CHALLENGES IN CURRENT A. CRYPTOCURRENCY

BLOCKCHAIN IMPLEMENTATIONS The famous cryptocurrency Bitcoin is one of the first imple-
One of the chief advantages of blockchain is its decentralized mentations of blockchain technology. It was proposed by a
nature. However, the fully decentralized nature of blockchain pseudonymous person (or group of people) named Satoshi
means that all data is available to all nodes; on a public Nakamoto. Bitcoin uses a public blockchain,for which trans-
blockchain, this means its available to the public, and on a actions are available to all the nodes in the peer-to-peer net-
consortium blockchain, all participants. This has the potential work. The transactions are created and written to a block.
to expose sensitive information of the parties involved, de- The new blocks are added to the blockchain based on the
pending on the data within each block. Access control meth- PoW consensus algorithm. Miners receive a portion of Bitcoin
ods may be imposed to ensure that only authorized personnel based on the amount of computation they spend to solve
have access to the data, but these methods usually require the the hashing problem. Overall, the average time to create
storage of passwords or biometric template, or possession of a new block is approximately 10 minutes. This amount of
token, which is contrary to the decentralized environment and time makes the possibility of a Distributed Denial of Service
making the traditional PKI unsuitable. If a user loses his/her (DDoS) attack almost infeasible.
private key, the privacy is exposed to dangerous threat. Such As Bitcoin’s blockchain is public, all information may be
mechanisms are often outside the paradigm, and are therefore exposed to the adversaries. The security of Bitcoin depends
also less effective. on the public-private key pair. Each address in a transaction is
Blockchain is utilized differently based on the application. associated with the public key. When a transaction is created,
Moreover, the difference in the data used in each the private key is used to sign without being disclosed to the
application differentiates the types of privacy. Hence, the public. The corresponding public key that is used to verify the
privacy-preserving techniques are not the same across various authority of this transaction. It is impossible to derive either
applications. There are some generic concepts that are the public key from the transaction’s address or the private
application-agnostic, but many advances are either specific key from the public key. This is a privacy issue for Bitcoin
to a field, or have not been considered for application in as in theory, without the private key to sign, Bitcoin cannot
other problem domains. In this article, we will look at the be sent with any transaction. However, if a user loses his/her
privacy-preserving techniques that have been devised in each private key to an adversary, the privacy of the user cannot be
field that applies blockchain technology. maintained. Bitcoin is also exposed to the 51-percent attack
in which the adversaries control a large amount of nodes in
III. APPLICATIONS OF BLOCKCHAIN the network. As a result, they take full control of the mining,
This section reviews the blockchain’s implementations to- which prevents other users from mining properly. On the other
wards different problem domains. In this survey, we catego- hand, with less than 51 percent of the whole network’s mining
rize the applications into the following types of blockchain capacity, the successful probability is much lower.
application area:
r Cryptocurrency B. DATA MANAGEMENT AND STORAGE
r Data Management and Storage One of the key drivers for Blockchain has been as a data
r E-voting storage mechanism, allowing for platform-wide auditing and
r IoT and Internet of Everything Ecosystems verification of data. This has several advantages: it provides

74 VOLUME 2, 2021
transparency for all participants; it makes data sharing more Ji et al. [31] sought to address the issue of location shar-
convenient whilst being simple to audit; it provides security ing in mobile medical services. They did so by proposing
defense techniques against the attack on centralized database. a blockchain-based multi-level privacy-preserving location
Thus, in this section, we look at how blockchain has been ap- sharing scheme. This scheme used Merkle trees with order-
plied to Health Records Management and Biometric Template preservation to achieve this. Evaluation on the computation
Protection. overhead of the scheme shows its feasibility, although it is yet
to be implemented.
Claiming that privacy issues in remote patient monitoring
1) HEALTH RECORDS MANAGEMENT might endanger a patient’s life, Dwivedi et al. [32] proposed
Electronic Health Records (EHR) storage is widely deployed to protect the users’ privacy in the blockchain network using
in many countries to provide the convenient sharing of in- a privacy-preserving ring signature scheme in addition to the
dividuals medical history. There are several advantages to double encryption of both data and the symmetric key. The
this; greater accountability, decreased administration burden, authors also stressed that this scheme is more appropriate to
and increased care, especially in the event of an accident or wearable IoT devices as it does not require heavy computa-
emergency situation. Due to its nature, blockchain and smart tional capability.
contracts [24] have been leveraged to accomplish this task. Zhang and Lin [33] aimed to improve diagnosis in e-
However, it also raises the issues about the sensitive health Health system by using two kinds of blockchain: A private
information of a patient’s being exposed to non-authorized blockchain to store the Personal Health Information (PHI)
personnel. This information could include (but is not limited and a consortium blockchain to store PHI’s indices. In this
to): medical history, treatments, location, and so on. There- scheme, privacy-preservation is obtained by encrypting all the
fore, privacy-preserving techniques must be incorporated to PHI data before storing in the private blockchain. Kuo and
protect the patient’s data. Machado proposed to solve the problem of privacy-preserving
Omar et al. [25] proposed MediBchain, a blockchain-based machine learning in healthcare with ModelChain, a frame-
electronic health record sharing framework in which the pa- work that incorporates private Blockchain network which con-
tient’s health data is encrypted and stored in a blockchain. tributes model parameter data without revealing the patients’
Sharing the data is dependent on the patient; only those who information [34].
are authorized by the patients can receive and decrypt data.
Dagher et al. [24] proposed Ancile, an Ethereum-based frame-
work to control the access to electronic health records with 2) BIOMETRIC TEMPLATE PROTECTION WITH
cryptographic techniques for security using smart contracts. BLOCKCHAIN TECHNOLOGY
Liu et al. [26] proposed BPDS, a blockchain-based privacy- A biometric template contains the biometric data recorded
preserving data sharing in which the original electronic med- from the enrollment process. This type of data is used to
ical records are stored in the cloud while the indices are be matched against newly retrieved biometric data to give a
stored in a consortium blockchain, resulting in not only the “match” or “non-match” decision. Biometric template protec-
mitigation of data leakage but also the prevention of arbitrary tion was devised such that if a protected template is compro-
data modification. On the other hand, Xu et al. [27] devised mised, it can be revoked and a new template can be estab-
Healthchain, a blockchain-based storage that is capable of lished. This means that the raw biometric data is safe. A noted
preserving the privacy of users by allowing them to add or current challenge in the biometric community is the security
revoke authorized doctors. Mohsin et al. [28] devised a fin- of biometric templates. Traditionally, biometric template data
ger vein verification framework that incorporates blockchain, is stored in a centralized database. However, if the database is
encryption, and steganography to protect a patient’s medical compromised, this template is lost to the hands of adversary,
information. This is unique in the sense that it integrates the who can use it to cross-authenticate other applications that use
use of steganography with other different security mecha- the same biometric. More importantly, biometric data (espe-
nisms in blockchain. Baqari and Barka [29] secured Electronic cially physiological type) is very limited and not as revocable
Health Records by incorporating biometric-based blockchain as a password would be. If a user loses his fingerprint to an
in which biometric is used to ensure not only the identity of a attacker, he or she might never be able to use it in future appli-
patient but also the recoverability of their access to the EHR cations. Hence, it is necessary to devise protection techniques
system. for the biometric template. There have been various methods
Similarly, Shen et al. [30] proposed a blockchain-based to achieve this goal. The most famous of these are cancelable
model to retrieve medical images while still preserving the templates and biometric cryptosystems. With its properties of
privacy of the patients. This model extracts feature vectors cryptographic security, decentralized nature, and unalterable
from each medical images are extracted to enable the cus- data transaction, blockchain has shown the potential to be the
tomization of the transaction design. Although this use-case is next biometric template protection method.
niche, it highlights the balance between privacy-preservation Goel et al. [35] presented a biometric recognition ar-
and utility in areas of e-health. chitecture that incorporated a private blockchain to extract

VOLUME 2, 2021 75
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

features. Matching is performed in a decentralized manner. e-voting as public trust is necessary while in reality, this is
The author showed that this architecture is able to reinforce rarely achievable. In addition, blockchain’s complexity might
the security of CNN-based model and biometric template by also be an obstacle when implementing e-voting for a region
implementing with different biometrics. The authors proposed with great population.
DeepRing [36] to protect CNN architecture from external Ayed [43] introduced a secure blockchain-based e-voting
adversaries using cryptography and blockchain technology. In system that achieves transparency by using open-source code.
addition, a tampering attack model is proposed to stress the This system does not allow a registration process, which
role that blockchain technology can play in the world of ma- makes it dependent on a separate database. Liu et al. [41] pro-
chine learning. This work enables blockchain to protect CNN posed an e-voting protocol that integrates blockchain without
models, including those that are used for biometric authenti- a trusted third party. Using blind signature [44], the authors
cation. Othman and Callahan [37] proposed Horcrux, a proto- claimed that this protocol is transparent to public and impor-
col that stores biometric credential in a decentralized manner tantly, maintains the anonymity of the voters as well as the
via blockchain using decentralized identifiers and documents privacy of the votes. Hjalmarsson et al. [45], in an on-going
(DIDs) developed by W3C Verifiable Claims. work, evaluated the use of blockhain as a service for an e-
Recently, Accquah et al. [38] proposed a method that pro- voting system. In details, the authors proposed a permissioned
tects the fingerprint templates with blockchain technology in blockchain that utilizes smart contracts to ensure the privacy
which fingerprint features are extracted, encrypted with AES, of the voters. However, the bridging problem has not been
and finally uploaded to a symmetrically distributed storage discussed in those works.
system called InterPlanetary File System (IPFS) [39] after
being split. The hash of the templates, separately to these, D. IOT AND INTERNET OF EVERYTHING ECOSYSTEMS
is stored on the Ethereum network. However, this work has
Internet of Things, commonly known as IoT, has attracted
three disadvantages that remain unaddressed; (1) biometrics
more and more interest in research due to its potentially wide
are known for their uncertainty, which is why they are in-
range of applications, and the promises it makes in increasing
compatible with conventional cryptography techniques. De-
flexibility, energy efficiency, and ease-of-use that such strong
spite this, the authors did not explain how they applied AES
integration can provide Things. With the escalating develop-
with fingerprints; (2) the process of matching has not been
ment of blockchain technology, this future is closer than ever.
discussed; and (3) although the cost and efficiency of storage
However, given the need for cloud integration, data collection
have been shown, the performance of the biometric authenti-
and remote analysis, the privacy of IoT users is potentially at
cation system has not been mentioned. This is an important
stake. The definition of IoT has expanded over time, not only
characteristic of a biometric authentication system.
incorporating user and home devices, but industrial systems.
This is often termed the Industrial Internet of Things (IIoT),
C. E-VOTING
or the Internet of Everything (IoE). This section is dedicated
Voting has played an important role in the development of
to review the efforts in research to protect users’ privacy in the
civilization as it allows the people of a country or territory
IoT, IIoT and IoE environments.
to choose their leaders. Voting is considered a cornerstone
of democracy. There have been significant achievements on
applying technology to voting, which if often referred to as 1) SECURITY ENHANCEMENT TECHNIQUES
Electronic voting or E-voting. There are numerous benefits to The security of IoT has been a consideration since its in-
e-voting; including precision, speed in counting, and logistic ception. The combination of low-cost, low-powered systems,
considerations [40], [41]. E-voting even is even more promis- difficulties in upgrades and fixes, and wireless networking,
ing when enabled by blockchain technology. The concept is has seen significant work outlining the potential security dan-
simple: Each voter has a electronic currency wallet. When the gers of IoT. Blockchain has been utilized to construct crypto-
voters vote for a candidate (or a number of candidates in some graphic protocols or in order to enhance the security in IoT
countries or regions), the electronic currency is transferred environments and ecosystems.
to the candidate(s)’ wallet [42]. However, due to the privacy In 2016, Kosba et al. [46] introduced Hawk, an Applica-
threats, e-voting has not been widely implemented. The po- tion Programming Interface (API) to help programmers write
tential disadvantages have been in the security of e-voting, privacy-preserving smart contracts code without the requir-
and ensuring privacy. Hence, there exists urgent necessity ing them to implement their own cryptography, lowering the
to devise privacy-preserving e-voting systems to ensure the complexity of implementing such systems. In 2017, Ouaddah
integrity of the voting process, whilst simultaneously ensuring et al. [47] introduced a blockchain-based framework that is
voter privacy. In e-voting, privacy refers to the identity of the capable of integrating privacy preservation into authorization
voters, the content of the vote, and their linkage, meaning management of access control. This API and framework have
that given a voter and a vote, one can not refer to one from enabled and expanded the use of blockchain to allow more
another. In addition, the security of the vote must be ensured applications to use blockchain technology without being con-
as its content cannot be tampered. The inherent privacy issues cerned about privacy leakage. In the same year, Kaaniche
from blockchain hinder the public adoption of blockchain and Laurent [48] presented a cryptographic technique using

76 VOLUME 2, 2021
blockchain technology to enable availability and accountabil- et al. [57] constructed a decentralized charging coordination
ity, whilst preserving privacy-preservation, in a a data usage mechanism to which ESUs can perform anonymous authenti-
auditing architecture. Recently, Shao et al. [49] achieved un- cation. In [58], Tan et al. addressed the problem of the privacy
linkability and anonymity for transactions of a legitimate user and security of the consumption and trading data using a
in a blockchain whereas malicious users can be identified blockchain-based energy scheduling model whose optimiza-
in a framework called AttriChain. Recently, Halpin intro- tion is separated into trivial scheduling problems, which are
duced Nym credentials as a novel method to deploy anony- then solved by consensus algorithm and smart contracts. Gai
mous authentication credentials to preserve user privacy in a et al. [59] aimed to protect the privacy of the users in a smart
blockchain environment [50]. grid by incorporating a permissioned blockchain with smart
Jiang et al. [51], [52] introduced Privacy-preserving Thin- contracts and edge computing in which group signatures and
client Authentication Scheme (PTAS) to improve the security covert channel authorization techniques are utilized. Keshk
for the users with limited storage IoT device by using Private et al. [60] devised a framework to provide data privacy and
Information Retrieval (PIR). In addition, thin-client data is security in smart power networks with two modules; The
protected against at most (m − 1) full node users’ collusion first being a two-level module that is dedicated to verify data
with the (m − 1) private PTAS. Evaluation on the computa- integrity using proof of work blockchain along with applying
tional and communication overhead of these two methods are a variational autoencoder to transform data, and the second
provided to show their potential. Recently, Patil et al. [53] pro- being an anomaly detection module for training and validating
posed a blockchain-based protocol that utilizes the Physically the output of the first module. This second module uses deep
Unclonable Function (PUF) model to achieve better authen- learning techniques. Experiments show its competitiveness
tication process whilst still preserving user privacy. In [54], against state-of-the-art techniques in protecting data and iden-
Jones et al. compared the use of trusted third parties, crypto- tify anomalies. Recently, Hy-Bridge [61] has been proposed
graphic techniques, and blockchain smart contracts and con- to not only help users secure their data against the service
cluded that each alone does not ensure privacy preservation. providers but also share credits with other users with the use
They implemented a hybrid approach combining the afore- of blockchain.
mentioned techniques to share geolocation data securely in a An electrical vehicle’s information may also be linked to its
trusted execution environment. With a similar aim of targeting owner, which reveals the privacy of an identity. Exposing this
the privacy when using location-based services, Qiu et al. [55] kind of information may lead to PII being leaked, including
a location privacy preserving approach that leverages multiple physical locations, times of movement, and information ex-
private blockchains to protect the users’ privacy without de- trapolated from these. Aiming to protect users privacy whilst
grading the service’s quality. This is still a proof-of-concept, charging Electrical Vehicles (EV) in a Vehicle-to-grid (V2G)
but offers a unique service if expanded. environment, Gao et al. [62] proposed a blockchain-based
payment method that is capable of sharing users’ data with-
out revealing the privacy by enabling registration with digital
2) SMART ENERGY signatures of a user and allowing auditing of payments by
Smart grids are an increasingly important form of Cyber- privileged users. On the other hand, Gabay et al. [63] intro-
Physical System (CPS) infrastructure, providing power for duced two approaches (token-based and Pederson Commit-
future IoT environments. In a smart grid, a smart meter ment Scheme-based) that incorporate zero knowledge proofs
is installed at each household to collect consumption data. with blockchain and smart contracts to protect the privacy of
This data enables adaptive and efficient power generation the charging EV when authenticating for charging.
and scheduling. Smart grids require the processing of smart
meter data to operate effectively. As a result, this data had the
potential to expose information that is potentially sensitive, 3) SMART TRANSPORTATION
including energy use patterns, and billing information. This Smart transportation is one of the key emerging use-cases
has significant second-order effects, as this data can be used to for IoT systems and infrastructure. The rise of smart vehi-
infer movement patterns and other private information. Failure cles and vehicular networks can apply some of the promises
to protect smart grid data may lead to disastrous consequences of IoT; increased safety, improved communications to as-
in real life. sist networks overall, and to provide timely updates to in-
There is emerging research that has sought to develop cidents such as accidents or road conditions. However, such
processes in smart energy for privacy preservation. Guan data also creates potential privacy risks, as it includes ve-
et al. [56] presented a method to protect smart grid user data hicle identity information, residual locations, vehicle move-
by separating the users into different groups, each of which ments, and messages transmitted across networks. A number
with a private blockchain for data recording. The user data of studies have been dedicated to apply blockchain and pri-
was held in a group protected from others in the same group vacy preservation in Vehicular Ad-hoc Networks (VANETs).
through the creation and use of pseudonyms associating to Sharma and Chakraborty [64] devised BlockAPP as a method
each individual. Seeking to address the problem of unco- to authenticate vehicles with their privacy being preserved
ordinated charging of Energy Storage Units (ESUs), Baza in the Internet of Vehicles (IoV). Lu et al. [65] proposed a

VOLUME 2, 2021 77
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

privacy-trust model for VANETs in which blockchain technol- respond with encrypted offers. The smart contract platform
ogy is employed to store the certificate and revocation trans- Ethereum was used to implement this method, which showed
parency. In 2019, the authors proposed a privacy-preserving feasibility in practice. Also working in privacy-enabled
authentication scheme in an extended blockchain environ- ridesharing, Li et al. [76] sought to address the problem
ment for VANETs [66]. Guehguih and Lu [67] proposed of cross-organizational data sharing between ridesharing
a dual blockchains as a solution for privacy-preserving au- service providers. They did this by devising CoRide, a
thentication in VANETs in which the private blockchain is privacy-preserving hailing service using blockchain-assisted
used for authentication while the public is utilized as event fog computing; a consortium blockchain to record the rides
messages manager. Li et al. [68] solved the problem of car- are constructed and smart contracts are utilized to pair
pooling users’ information being disclosed in fog computing riders with drivers. An Ethereum network was also used to
by incorporating blockchain-assisted vehicular fog computing implement this system. However, in this work, the assumption
that is capable of preserving the users’ private information. is that all users adopt the same data formats for matching and
Feng et al. [69] implemented a blockchain-assisted privacy- payment. This is rarely the practice in reality, as rideshare
preserving authentication system (BPAS) in VANETs to ver- platforms have their own technology stacks and formats.
ify the messages being transmitted in the network. Security
analysis shows that this framework is able to achieve condi-
tional privacy preservation. 4) SMART AGRICULTURE
To solve the problem of user anonymization and lack of mo- Agriculture has played a crucial role in the development of
tivation to forward announcements in Vehicular Social Net- any civilization. With the development of new technologies,
works (VSNs), Li et al. [70] proposed CreditCoin, a privacy- agriculture has promising ways to increase yields, quality and
preserving blockchain-based announcement network with product shelf life. There are also changes in pressure from
incentives, where announcements can be sent via an anony- consumers, who increasingly want to know more about their
mous vehicular announcement aggregation protocol. Despite food, including the origins of ingredients, details of the pro-
the anonymous nature, the identity of malicious users can cesses different foodstuffs have gone through over its develop-
be tracked with the system. Pu et al. proposed a privacy- ment, and the chemicals and preservatives it has been exposed
preserving scheme that is capable of concealing a vehicle’s to. Smart Agriculture has been leveraged to achieve these
identity in a VSN [71]. To prevent tampering, the data of goals as it integrates traditional agriculture with advanced
the VSN is stored in a blockchain and Practical Byzantine technologies such as IoT. One of the recent trends for smart
Fault Tolerance (PBFT) is employed for the purpose of con- agriculture is the use of blockchain to ensure the transparency
sensus. In [72], the authors proposed to combine a consor- of food data. This needs to be balanced with the obvious
tium blockchain with a privacy-preserving location-based ser- commercial privacy requirements, as much of the information
vice to address the issue of man-in-the-middle-attack and the that can be collected with IoT devices is unique to each pro-
requirement of certificate authority in communicating with ducer. Measurements such as soil moisture, temperatures and
limited-resources smart vehicles in a VSN. A lightweight fertilisers used are commercially sensitive. Affording relevant
certificate authority is devised in addition to the use of the information to consumers and buyers without leaking sensi-
privacy-preserving location-based service protocol, PPVC. tive proprietary data is a unique challenge.
The experimental evaluation and security analysis show the In 2018, Kim et al. [77] introduced Harvest Network, a
efficiency of this method in a VSN environment. farm-to-fork food traceability application that leverages smart
Smart transport has aspects far beyond the individual vehi- contracts and GS1 message exchanging standards for IoT
cle networks, and there are several other aspects of the field devices. However, this framework was only theoretical as
that have benefited from blockchain technology in a privacy- there was no prototype to evaluate the effectiveness at the
preserving manner to provide convenience and security. time. Caro et al. [4] devised AgriBlockIoT, a food traceabil-
Hu et al. [73] proposed a smart parking framework to ity solution that is based on fully centralized blockchain. In
protect the privacy of users. It did this with a blockchain to details, the authors achieved traceability by deploying two
remove the need for a trusted third party. Similarly, Amiri blockchains on Ethereum and Hyperledger Sawtooth, respec-
et al. [74] applied a blockchain with PIR to construct a tively. Performance of the system on latency, CPU, network
smart parking system that does not disclose drivers’ sensitive usage have been provided for a comparison of pros and cons.
information. Lin et al. [78] devised a tamper-proof food traceability system
Interested in ridesharing, Baza et al. [75] claimed that based on blockchain to monitor the food’s lifespan in smart
a myriad of problems caused by the traditional central agriculture. This system integrates trustworthy verification for
third-party ridesharing methods, including privacy concerns, blockchain and is able to work with low-power wide-area
could be resolved using their proposed public Blockchain in network (LPWAN) IoT system. However, no implementation
which drivers are allowed to offer ridesharing services without and experiment results have been discussed. Salah et al. [79]
having to rely on a central provider. On the client’s side, proposed a smart contract-based approach to perform business
sensitive information such as locations, date, and time are transactions and traceability for soy bean in the agriculture
preserved by sending cloaked requests, to which the drivers supply chain. All transactions from every party involved in

78 VOLUME 2, 2021
the process are recorded in the blockchain with links to an have proposed a privacy-preserving mobile crowdsensing ar-
IPFS [39]. chitecture for a smart framing application. In this architec-
ture, a perturbation-based privacy mechanism is developed
in conjunction with a trusted third party to ensure user pri-
E. OTHER FIELDS vacy in the mobile crowdsensing environment. In general, no
The applicability of blockchain technology is still growing as adequate and dedicated security/privacy data security infras-
it has been utilized in an increasing number of fields, including tructure for smart farming. We observe that PPBS is an ideal
access control, cross-domain data sharing, and data integrity security technology platform for smart farming as smart farm-
in IoT environments. Due to the huge amount of data to be ing would need to securely store large amount of timestamped
used with blockchain, it is critical to use privacy-preserving data which could be accessed by third parties. Also, the smart
aspects to ensure the security. Makhdoom et al. [80] pro- contract technology can help remove the cost of middleman in
posed a method to share multiple types of data in a smart finding individual goods suppliers to collectively fulfil a big
city environment based on blockchain technology using multi- product order placed by a retailer. Therefore, we propose the
channel approach with the capability to protect users’ pri- Privacy Preserving Smart Agriculture Framework (PPSAF),
vacy. Zhao et al. [81] presented a privacy-preserving software a PPBS-based secure data framework for the smart farming.
update protocol using blockchain technology with incentive. The proposed framework is shown in Fig. 1.
Zhao et al. [82] addressed the problem of remote data in- The PPSAF framework is comprised of the following com-
tegrity checking for IoT Systems with a privacy-preserving ponents; field sensor networks, an aggregator, gateways, and a
blockchain scheme. Le and Mutka [83] introduced CapChain, consortium blockchain. As outlined in Fig. 1, the system oper-
a blockchain-based access control framework for users to ates as follows: field sensor networks continuous collect data
share their access rights to IoT devices in public with privacy from crop fields, and the aggregator sends collected sensor
preservation. Lu et al. [84] applied blockchain to the problem network to the gateway of the farmer. The farmer stores the
of authentication in cross-organizational data sharing without data to its own private ledger (private channel) in the cloud.
the existence of a trusted third party. Targeting the users’ This data may include crop information, site information,
privacy in recommendation systems, Casino and Patsakis [85] fertilizer distribution, soil readings, and other relevant infor-
devised a Collaborative Filtering system based on blockchain mation. Farmers, banks, logistics and retailer form a consor-
to construct a decentralized architecture that is based on local- tium blockchain built on a Hyperledger Fabric Platform. Data,
ity sensitive hashing classification. including business transactions are encrypted, which can be
In [86], the authors discussed three location-related vul- accessed and verified among the consortium members, once
nerabilities in the traditional crowdsensing system. They pro- authorized. Major players such as banks and retailers form
posed a privacy-preserving crowdsensing system based on the initial system and other potential participants must register
blockchain technology with incentive to address these issues first, subject to the approval from super nodes. The proposed
as well as motivate the workers to complete the task. Chen framework provides two major applications for smart farming:
et al. [87] described DEPLEST as a portable-device- (1) Secure Data Infrastructure for Data Storage and Access,
compatible method to protect the social network users’ pri- and (2) Smart Contract for Business Bidding. Each of these is
vacy in which a modified blockchain model to secure their discussed separately.
private information and passes non-private information for
processing. In addition, a consensus protocol for blockchain A. SECURE DATA INFRASTRUCTURE FOR DATA
ledger maintenance that is Byzantine-Fault Tolerant (BTF) is STORAGE AND ACCESS
also proposed with experimental results to show its feasibility The power of smart farming lies in the smart use of data for ac-
when compared with the PoW and PoS methods. curate farming. However, data such as fertilizer density in the
field, field temperate and moisture, are commercially sensitive
data and therefore must be protected. The data could also be
IV. PPSAF, A NOVEL FRAMEWORK FOR PRIVACY used by the trusted third party in providing big data analysis
PRESERVATION IN SMART AGRICULTURE service. The PPSAF framework provides the following solu-
Although there are many successful applications of PPBS tion: In the local sensor network, field data (up to the aggre-
(such as PPBS-based transportation systems and PPBS based gator) are encrypted with a lightweight cryptography mecha-
e-health systems, for example), its application to smart farm- nism. The data will then be forwarded to the farmer’s gateway
ing is largely an unexplored territory. It is estimated that we which is installed in the farm. Upon receiving the encrypted
need to produce 70% more food in 2025 than it did in 2006 in data forwarded by the aggregator, the gateway decrypts the
order to feed the growing population [95]. Therefore, smart data, compresses them into corresponding time-interval (e.g.,
farming has been heralded as the future of agriculture. hourly) blocks. The farmer encrypts this block of data with a
Data infrastructure is a foundational technology for smart stronger cryptographic scheme, signs the encrypted data block
farming and agriculture, whereby data is used to optimize which will be placed in the farmer’s private blockchain in the
farming efficiency and quality. Security and privacy issues can cloud. A signed transaction notice will be broadcast in the
hinder its large-scale deployment. Recently, Huning et al. [96] consortium blockchain. This transaction notice will contain

VOLUME 2, 2021 79
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

FIGURE 1. PPBS-based Framework For Smart agriculture.

the description of the data block including the pointer to data retailer agent RA would like to place an order for certain
block stored in the farmer’s private blockchain channel in the goods, e.g., grapes, from farmers. For convenience, we as-
cloud. The consortium blockchain will store this transaction sume relevant transactions are broadcast over the Consortium
notice once the involved data block and transaction notice Blockchain, signed and are validated by the Blockchain, un-
have been confirmed by the Consortium via majority voting. less specified otherwise. After the broadcast has occurred, the
As the value of a data block is associated with the time, e.g., negotiation process is conducted out of band. Goods orders
field temperature in the morning, and evening etc., each block are as follows:
should be encrypted with a different session key if symmetric Step 1: RA fetches the membership ID list of the Consor-
encryption is used. Or different data block is equipped with tium Blockchain and selects a group S that consists of targeted
a different access control policy so that access to the data is members, e.g., grape farmers.
based on the block-level. Step 2: RA constructs an order for goods in a smart contract
Data Access: format specifying legal terms/conditions, parameters of goods
r Owner access: has full access such as quality, minimum individual supply quantity thresh-
r Access from other consortium members: other members old, minimum and maximum total supply quantity, banking
have access to the encrypted data block in the owner’s institutions, and date of delivery etc...
private blockchain channel in the cloud for the verifica- Step 3: RA broadcast the signed and encrypted smart con-
tion purpose. Access to the content of a specific data tract to the group S via Identity-based broadcast encryption
block must get permission from the data owner. The schemes.
negotiation of the permission including the decryption Step 4: Farmers in the group S can decrypt the smart con-
key distribution is conducted out of band. tract. Those who are interested in supplying the goods and are
r Third party access: No public access to the owner’s pri- satisfied with the terms and conditions set in the smart contract
vate channel. Access to the data must be negotiated with will broadcast their signed agreements that are encrypted with
the owner via out of band mechanisms. RA’s public key.
Step 5: RA’s smart contract will periodically check trans-
B. SMART CONTRACT FOR BUSINESS BIDDING mitted agreements and the smart contract will be triggered
Secure automated business biding and trading order process- once the condition, such as total supply volume threshold, is
ing are another major application of smart framing. There is met.
no adequate and dedicated framework for this process. The Step 6: Similar to Step 3, the bank broadcast the payment
PPSAF framework can also be applied to addressing issue. transaction receipts for RA’s payment/deposit to individual
The proposed solution is illustrated as follows: Suppose a farmer/supplier Fi which are encrypted with the group (RA,

80 VOLUME 2, 2021
Fi ). This ensures only the payment recipient Fi and the payer privacy. In addition, due to the fact that biometric authentica-
RA can decrypt the receipts. At this stage, we assign the tion systems have provided a more convenient way to control
bank to issuing payment transaction evidence only and do the access in IoT, the protection of biometric templates in IoT
not include a smart contract triggering the bank’s automated environment is critical since unlike database where it is stored
payment. This is because existing banking systems have either in a centralized manner, blockchain has decentralized the IoT
in-person security check for large sum payment or Internet environment. Therefore, decentralized biometric authentica-
banking, which offers more security control. tion systems with template protection mechanism should be
Step 7: Individual supplier Fi will arrange logistics for the the next interesting area to which researchers pay attention.
goods delivery. Similar to the RA, logistics companies can
also deploy smart contracts for the order to arrange delivery VI. CONCLUSION
for a set of suppliers. As blockchain has found its way to be a useful tool in real
This protocol can be tailored to accommodate bidding pro- life thanks to its decentralized manner and robustness against
cess. For example, multiple logistics companies can place data manipulation, the privacy of the involved parties as well
different orders, which can only be viewed by targeted sup- as data transparency is still an open question. In order to
pliers. Farmers can choose their preferred contracts securely fully leverage the power of blockchain technology without
and independently. compromising the security, techniques to ensure the privacy
ought to be implemented.
In this survey, we have discussed the current challenges
V. FUTURE TRENDS FOR PRIVACY PRESERVATION WITH in privacy protection when applying Blockchain in differ-
BLOCKCHAIN TECHNOLOGIES ent fields. Based on that, we have provided two layers
As we can see, blockchain has already expanded its appli- of categorization for privacy-preserving blockchain appli-
cability to more and more areas. As a result, more fields cations by first reviewing some key advances in different
should appear with privacy protection in the existing list of blockchain’s applicability areas, then categorized the types
blockchain applications enabled. There are several domains in of privacy accordingly. Importantly, we proposed PPSAF, a
which there is both a need for transparency and tamper-proof. novel Privacy-Preserving Smart Agriculture Framework based
Let us take E-voting for example. E-voting requires all the on blockchain to further explore this technology’s potential
processes to be transparent and accurate. However, the pro- in smart farming and smart agriculture. The future trends for
posed E-voting schemes have not resolved the privacy require- privacy protection in blockchain have also been discussed, as
ments to ensure the anonymity of the voters and especially the we expect advances in the areas of blockchain and privacy-
bridging from voters to vote data. Therefore, E-voting is very preservation to create new opportunities in e-voting, smart
likely to attract interest with the application of blockchain farming and agriculture, smart cities, and smart contract-
with privacy-preserving techniques. In addition, given that enabled applications in the near future. As a result, the de-
more applications have been smart-contracts-enabled, there is mand for a better security while maintaining the transparency
a potential for both blockchain and smart contracts to assist in and decentralization will only increase, and technologies such
this space. as blockchain will likely underpin these advances.
There are several architectural areas of future work aspects
that, once developed, will provide significant benefit. There REFERENCES
is a need for a greater number of reference blockchain and [1] N. Hackius and M. Petersen, “Blockchain in logistics and supply chain:
smart contracts architectures for different purposes, preferably Trick or treat?,” in Proc. Digitalization Supply Chain Manage. Logis-
tics: Smart Digit. Solutions Ind. 4.0 Environ. Proc. Hamburg Int. Conf.
in a field-agnostic manner. This could provide templates and Logistics, vol. 23. Berlin: epubli GmbH, 2017, pp. 3–18.
reference architectures that assist across many fields, allow- [2] S. Saberi, M. Kouhizadeh, J. Sarkis, and L. Shen, “Blockchain technol-
ing researchers and developers to implement best-of-breed ogy and its relationships to sustainable supply chain management,” Int.
J. Prod. Res., vol. 57, no. 7, pp. 2117–2135, 2019.
adaptations with confidence that their implementation is fit [3] N. Kshetri, “1 blockchain’s roles in meeting key supply chain manage-
for purpose. This would not work across all fields, as each ment objectives,” Int. J. Inf. Manage., vol. 39, pp. 80–89, 2018.
implementation may require unique characteristics. [4] M. P. Caro, M. S. Ali, M. Vecchio, and R. Giaffreda, “Blockchain-based
traceability in agri-food supply chain management: A practical imple-
Given the continuing research and development in IoT, mentation,” in Proc. IoT Vertical Topical Summit Agriculture-Tuscany.,
there is an increasing need for more work to integrate 2018, pp. 1–4.
Blockchain and smart contract paradigms into these emerging [5] R. Cole, M. Stevenson, and J. Aitken, “Blockchain technology: Impli-
cations for operations and supply chain management,” Supply Chain
ecosystems. Importantly, fuelled by success, the promises of Manage. An Int. J., vol. 24, no. 4, pp. 469–483, 2019.
IoT are closer than ever, with multiple paradigms that are able [6] D. Shakhbulatov, J. Medina, Z. Dong, and R. Rojas-Cessa, “How
to communicate with one another, resulting in more consensus blockchain enhances supply chain management: A survey,” IEEE Open
J. Comput. Soc., vol. 1, pp. 230–249, 2020.
algorithms. This will lead to an increase in IoT ecosystems [7] T.-T. Kuo, H.-E. Kim, and L. Ohno-Machado, “Blockchain distributed
that actively use consensus algorithms and blockchain. Hence, ledger technologies for biomedical and health care applications,” J.
there is a necessity to ensure the privacy of systems, devices, Amer. Med. Informat. Assoc., vol. 24, no. 6, pp. 1211–1220, 2017.
[8] G. Drosatos and E. Kaldoudi, “Blockchain applications in the biomedi-
data and entities. Security techniques and protocols are also cal domain: A scoping review,” Comput. Struct. Biotechnol. J., vol. 17,
expected to bloom, providing more various ways to ensure the pp. 229–240, 2019.

VOLUME 2, 2021 81
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

[9] M. Johnson, M. Jones, M. Shervey, J. T. Dudley, and N. Zimmerman, [33] A. Zhang and X. Lin, “Towards secure and privacy-preserving data
“Building a secure biomedical data sharing decentralized app (DAPP): sharing in e-health systems via consortium blockchain,” J. Med. Syst.,
Tutorial,” J. Med. Internet Res., vol. 21, no. 10, 2019, Art. no. e13601. vol. 42, no. 8, 2018, Art. no. 140.
[10] A. F. Hussein, A. K. ALZubaidi, Q. A. Habash, and M. M. Jaber, “An [34] T.-T. Kuo and L. Ohno-Machado, “Modelchain: Decentralized privacy-
adaptive biomedical data managing scheme based on the blockchain preserving healthcare predictive modeling framework on private
technique,” Appl. Sci., vol. 9, no. 12, 2019, Art. no. 2494. blockchain networks,” 2018, arXiv:1802.01746.
[11] J. H. Park and J. H. Park, “Blockchain security in cloud computing: Use [35] A. Goel, A. Agarwal, M. Vatsa, R. Singh, and N. Ratha, “Securing
cases, challenges, and solutions,” Symmetry, vol. 9, no. 8, 2017, Art. no. cnn model and biometric template using blockchain,” IEEE Int. Conf.
164. Biometrics Theory, Appl. Syst., 2019, pp. 1–6.
[12] Y. Zhang, R. H. Deng, X. Liu, and D. Zheng, “Blockchain based effi- [36] A. Goel, A. Agarwal, M. Vatsa, R. Singh, and N. Ratha, “
cient and robust fair payment for outsourcing services in cloud comput- Deepring: Protecting deep neural network with blockchain,” in
ing,” Inf. Sci., vol. 462, pp. 262–277, 2018. Proc. IEEE Conf. Comput. Vis. Pattern Recognit. Workshops, 2019,
[13] O. Jacobovitz, “Blockchain for identity management,” The Lynne pp. 2821–2828.
William Frankel Center Comput. Sci. Dept. Comput. Sci. Beer Sheva: [37] A. Othman and J. Callahan, “The horcrux protocol: A method for de-
Ben-Gurion University, 2016. centralized biometric-based self-sovereign identity,” in Proc. Int. Joint
[14] M. Kuperberg, “Blockchain-based identity management: A survey from Conf. Neural Netw., 2018, pp. 1–7.
the enterprise and ecosystem perspective,” IEEE Trans. Eng. Manag., [38] M. A. Acquah, N. Chen, J.-S. Pan, H.-M. Yang, and B. Yan, “Securing
vol. 67, no. 4, pp. 1008–1027, Nov. 2020. fingerprint template using blockchain and distributed storage system,”
[15] A. V. Ertemel, “Implications of blockchain technology on marketing,” Symmetry, vol. 12, no. 6, 2020, Art. no. 195.
J. Int. Trade, Logistics Law, vol. 4, no. 2, pp. 35–44, 2018. [39] “Interplanetary file system,” Accessed: Dec. 1, 2020. [Online]. Avail-
[16] D. Liu, A. Alahmadi, J. Ni, X. Lin, and X. Shen, “Anonymous reputa- able: https://ipfs.io/
tion system for iiot-enabled retail marketing atop pos blockchain,” IEEE [40] D. A. Gritzalis, “Principles and requirements for a secure e-voting
Trans. Ind. Informat., vol. 15, no. 6, pp. 3527–3537, Jun. 2019. system,” Comput. Secur., vol. 21, no. 6, pp. 539–556, 2002.
[17] A. Rejeb, J. G. Keogh, and H. Treiblmaier, “How blockchain tech- [41] Y. Liu and Q. Wang, “An e-voting protocol based on blockchain,” IACR
nology can benefit marketing: Six pending research areas,” Front. Cryptol. Eprint Arch., vol. 2017, 2017, Art. no. 1043.
Blockchain, vol. 3, pp. 1–12, 2020. [42] N. Kshetri and J. Voas, “Blockchain-enabled e-voting,” IEEE Softw.,
[18] I. Antoniadis, K. Spinthiropoulos, and S. Kontsas, “Blockchain appli- vol. 35, no. 4, pp. 95–99, Jul./Aug. 2018.
cations in tourism and tourism marketing: A short review,” in Strategic [43] A. B. Ayed, “A conceptual secure blockchain-based electronic voting
Innovative Marketing and Tourism. Berlin, Germany: Springer, 2020, system,” Int. J. Netw. Secur. Its Appl., vol. 9, no. 3, pp. 1–9, 2017.
pp. 375–384. [44] D. Chaum, “Blind signature system,” in Advances in Cryptology. Berlin,
[19] T. Hwang, Subprime Attention Crisis: Advertising and the Time Bomb Germany: Springer, 1984, pp. 153–153.
At the Heart of the Internet. New York, NY, USA: Farrar Straus Giroux, [45] F. Hjalmarsson, G. K. Hreidharsson, M. Hamdaqa, and G. Hjalmtysson,
2020. “Blockchain-based e-voting system,” in Proc. IEEE 11th Int. Conf.
[20] T. Blake, C. Nosko, and S. Tadelis, “Consumer heterogeneity and paid Cloud Comput., 2018, pp. 983–986.
search effectiveness: A large-scale field experiment,” Econometrica, [46] A. Kosba, A. Miller, E. Shi, Z. Wen, and C. Papamanthou,
vol. 83, no. 1, pp. 155–174, 2015. “Hawk: The blockchain model of cryptography and privacy-preserving
[21] B. Shapiro, G. J. Hitsch, and A. Tuchman, “Generalizable and robust tv smart contracts,” in Proc. IEEE Symp. Secur. Privacy., 2016,
advertising effects,” NBER Working Paper, 2020, Paper w27684. pp. 839–858.
[22] D. L. Chaum, Computer Systems Established, Maintained and Trusted [47] A. Ouaddah, A. Abou Elkalam, and A. A. Ouahman, “Towards a novel
by Mutually Suspicious Groups. Electronics Research Laboratory, Univ. privacy-preserving access control model based on blockchain technol-
California, California, USA, 1979. ogy in IoT,” in Europe and MENA Cooperation Advances in Infor-
[23] D. Bayer, S. Haber, and W. S. Stornetta, “Improving the efficiency and mation and Communication Technologies. Berlin, Germany: Springer,
reliability of digital time-stamping,” in Sequences Ii. Berlin, Germany: 2017, pp. 523–533.
Springer, 1993, pp. 329–334. [48] N. Kaaniche and M. Laurent, “A blockchain-based data usage auditing
[24] G. G. Dagher, J. Mohler, M. Milojkovic, and P. B. Marella, “Ancile: architecture with enhanced privacy and availability,” in Proc. IEEE 16th
Privacy-preserving framework for access control and interoperability of Int. Symp. Netw. Comput. Appl., 2017, pp. 1–5.
electronic health records using blockchain technology,” Sustain. Cities [49] W. Shao, C. Jia, Y. Xu, K. Qiu, Y. Gao, and Y. He, “Attrichain: Decen-
Soc., vol. 39, pp. 283–297, 2018. tralized traceable anonymous identities in privacy-preserving permis-
[25] A. Al Omar, M. S. Rahman, A. Basu, and S. Kiyomoto, “Medibchain: sioned blockchain,” Comput. Secur., 2020, Art. no. 102069.
A blockchain based privacy preserving platform for healthcare data,” [50] H. Halpin, “Nym credentials: Privacy-preserving decentralized identity
in Proc. Int. Conf. Secur., Privacy Anonymity Computation, Commun. with blockchains,” in Proc. Crypto Valley Conf. Blockchain Technol.,
Storage. Berlin, Germany: Springer, 2017, pp. 534–543. 2020, pp. 56–67.
[26] J. Liu, X. Li, L. Ye, H. Zhang, X. Du, and M. Guizani, “Bpds: A [51] W. Jiang, H. Li, G. Xu, M. Wen, G. Dong, and X. Lin, “A privacy-
blockchain based privacy-preserving data sharing for electronic medical preserving thin-client scheme in blockchain-based PKI,” in Proc. IEEE
records,” in Proc. IEEE Global Commun. Conf., 2018, pp. 1–6. Glob. Commun. Conf., 2018, pp. 1–6.
[27] J. Xu et al., “Healthchain: A blockchain-based privacy preserving [52] W. Jiang, H. Li, G. Xu, M. Wen, G. Dong, and X. Lin, “PTAS: Privacy-
scheme for large-scale health data,” IEEE Internet Things J., vol. 6, preserving thin-client authentication scheme in blockchain-based PKI,”
no. 5, pp. 8770–8781, Oct. 2019. Future Gener. Comput. Syst., vol. 96, pp. 185–195, 2019.
[28] A. Mohsin et al., “Based blockchain-PSO-AES techniques in finger [53] A. S. Patil, R. Hamza, A. Hassan, N. Jiang, H. Yan, and J. Li, “Efficient
vein biometrics: A novel verification secure framework for patient privacy-preserving authentication protocol using pufs with blockchain
authentication,” Comput. Standards Interfaces, vol. 66, 2019, Art. no. smart contracts,” Comput. Secur., vol. 97, 2020, Art. no. 101958.
103343. [54] M. Jones, M. Johnson, M. Shervey, J. T. Dudley, and N. Zimmerman,
[29] M. A. Baqari and E. Barka, “Biometric-based blockchain ehr sys- “Privacy-preserving methods for feature engineering using blockchain:
tem (bbehr),” in Proc. Int. Wireless Commun. Mobile Comput., 2020, Review, evaluation, and proof of concept,” J. Med. Internet Res., vol. 21,
pp. 2228–2234. no. 8, 2019, Art. no. e13600.
[30] M. Shen, Y. Deng, L. Zhu, X. Du, and N. Guizani, “Privacy-preserving [55] Y. Qiu, Y. Liu, X. Li, and J. Chen, “A novel location privacy-preserving
image retrieval for medical IoT systems: A blockchain-based ap- approach based on blockchain,” Sensors, vol. 20, no. 12, 2020, Art. no.
proach,” IEEE Netw., vol. 33, no. 5, pp. 27–33, Sep.–Oct. 2019. 3519.
[31] Y. Ji, J. Zhang, J. Ma, C. Yang, and X. Yao, “BMPLS: Blockchain- [56] Z. Guan et al., “Privacy-preserving and efficient aggregation based
based multi-level privacy-preserving location sharing scheme for tele- on blockchain for power grid communications in smart communities,”
care medical information systems,” J. Med. Syst., vol. 42, no. 8, 2018, IEEE Commun. Mag., vol. 56, no. 7, pp. 82–88, Jul. 2018.
Art. no. 147. [57] M. Baza, M. Nabil, M. Ismail, M. Mahmoud, E. Serpedin, and M.
[32] A. D. Dwivedi, G. Srivastava, S. Dhar, and R. Singh, “A decentralized A. Rahman, “Blockchain-based charging coordination mechanism for
privacy-preserving healthcare blockchain for IoT,” Sensors, vol. 19, smart grid energy storage units,” in Proc. IEEE Int. Conf. Blockchain,
no. 2, 2019, Art. no. 326. 2019, pp. 504–509.

82 VOLUME 2, 2021
[58] S. Tan, X. Wang, and C. Jiang, “Privacy-preserving energy scheduling [79] K. Salah, N. Nizamuddin, R. Jayaraman, and M. Omar, “Blockchain-
for escos based on energy blockchain network,” Energies, vol. 12, no. 8, based soybean traceability in agricultural supply chain,” IEEE Access,
2019, Art. no. 1530. vol. 7, pp. 73295–73305, 2019.
[59] K. Gai, Y. Wu, L. Zhu, L. Xu, and Y. Zhang, “Permissioned blockchain [80] I. Makhdoom, I. Zhou, M. Abolhasan, J. Lipman, and W. Ni,
and edge computing empowered privacy-preserving smart grid net- “PrivySharing: A blockchain-based framework for privacy-preserving
works,” IEEE Internet Things J., vol. 6, no. 5, pp. 7992–8004, and secure data sharing in smart cities,” Comput. Secur., vol. 88, 2020,
Oct. 2019. Art. no. 101653.
[60] M. Keshk, B. Turnbull, N. Moustafa, D. Vatsalan, and K.-K. R. Choo, [81] Y. Zhao, Y. Liu, A. Tian, Y. Yu, and X. Du, “Blockchain based
“A privacy-preserving-framework-based blockchain and deep learning privacy-preserving software updates with proof-of-delivery for internet
for protecting smart power networks,” IEEE Trans. Ind. Informat., of things,” J. Parallel Distrib. Comput., vol. 132, pp. 141–149, 2019.
vol. 16, no. 8, pp. 5110–5118, Aug. 2020. [82] Q. Zhao, S. Chen, Z. Liu, T. Baker, and Y. Zhang, “Blockchain-based
[61] M. Daghmehchi Firoozjaei, A. Ghorbani, H. Kim, and J. Song, “Hy- privacy-preserving remote data integrity checking scheme for IoT in-
bridge: A hybrid blockchain for privacy-preserving and trustful energy formation systems,” Inf. Process. Manage., vol. 57, no. 6, 2020, Art. no.
transactions in internet-of-things platforms,” Sensors, vol. 20, no. 3, p. 102355.
928, 2020. [83] T. Le and M. W. Mutka, “Capchain: A privacy preserving access control
[62] F. Gao, L. Zhu, M. Shen, K. Sharif, Z. Wan, and K. Ren, “A blockchain- framework based on blockchain for pervasive environments,” in Proc.
based privacy-preserving payment mechanism for vehicle-to-grid net- IEEE Int. Conf. Smart Comput., 2018, pp. 57–64.
works,” IEEE Netw., vol. 32, no. 6, pp. 184–192, Nov./Dec. 2018. [84] P. J. Lu, L.-Y. Yeh, and J.-L. Huang, “An privacy-preserving cross-
[63] D. Gabay, K. Akkaya, and M. Cebe, “Privacy-Preserving authenti- organizational authentication/authorization/accounting system using
cation scheme for connected electric vehicles using blockchain and blockchain technology,” in Proc. IEEE Int. Conf. Commun., 2018,
zero knowledge proofs,” IEEE Trans. Veh. Technol., vol. 69, no. 6, pp. 1–6.
pp. 5760–5772, Jun. 2020. [85] F. Casino and C. Patsakis, “An efficient blockchain-based privacy-
[64] R. Sharma and S. Chakraborty, “Blockapp: Using blockchain for au- preserving collaborative filtering architecture,” IEEE Trans. Eng.
thentication and privacy preservation in iov,” in Proc. IEEE Globecom Manag., vol. 67, no. 4, pp. 1501–1513, 2019.
Workshops, 2018, pp. 1–6. [86] M. Yang, T. Zhu, K. Liang, W. Zhou, and R. H. Deng, “A blockchain-
[65] Z. Lu, W. Liu, Q. Wang, G. Qu, and Z. Liu, “A privacy-preserving trust based location privacy-preserving crowdsensing system,” Future Gener.
model based on blockchain for vanets,” IEEE Access, vol. 6, pp. 45 655– Comput. Syst., vol. 94, pp. 408–418, 2019.
45 664, 2018. [87] Y. Chen, H. Xie, K. Lv, S. Wei, and C. Hu, “Deplest: A blockchain-
[66] Z. Lu, Q. Wang, G. Qu, H. Zhang, and Z. Liu, “A blockchain-based based privacy-preserving distributed database toward user behaviors in
privacy-preserving authentication scheme for vanets,” IEEE Trans. social networks,” Inf. Sci., vol. 501, pp. 100–117, 2019.
Very Large Scale Integr. (VLSI) Syst., vol. 27, no. 12, pp. 2792–2801, [88] D. Pawade, A. Sakhapara, A. Badgujar, D. Adepu, and M. Andrade,
Dec. 2019. “Secure online voting system using biometric and blockchain,” in Data
[67] B. Guehguih and H. Lu, “Blockchain-based privacy-preserving authen- Management Analytics and Innov. Berlin, Germany: Springer, 2020,
tication and message dissemination scheme for vanet,” in Proc. 5th Int. pp. 93–110.
Conf. Syst., Control Commun., 2019, pp. 16–21. [89] E. Akbari, Q. Wu, W. Zhao, H. R. Arabnia, and M. Q. Yang, “From
[68] M. Li, L. Zhu, and X. Lin, “Efficient and privacy-preserving carpool- blockchain to internet-based voting,” in Proc. Int. Conf. Comput. Sci.
ing using blockchain-assisted vehicular fog computing,” IEEE Internet Comput. Intell., 2017, pp. 218–221.
Things J., vol. 6, no. 3, pp. 4573–4584, Jun. 2019. [90] S. Panja and B. K. Roy, “A secure end-to-end verifiable e-voting system
[69] Q. Feng, D. He, S. Zeadally, and K. Liang, “Bpas: Blockchain-assisted using zero knowledge based blockchain,” IACR Cryptol. Eprint Arch.,
privacy-preserving authentication system for vehicular ad hoc net- vol. 2018, 2018, Art. no. 466.
works,” IEEE Trans. Ind. Informat., vol. 16, no. 6, pp. 4146–4155, [91] R. Nimje and D. Bhalerao, “Blockchain based electronic voting system
Jun. 2020. using biometric,” in Proc. Int. Conf. Sustain. Commun. Netw. Appl.
[70] L. Li et al., “Creditcoin: A privacy-preserving blockchain-based in- Berlin, Germany: Springer, 2019, pp. 746–754.
centive announcement network for communications of smart vehi- [92] T. Roopak and R. Sumathi, “Electronic voting based on virtual id of
cles,” IEEE Trans. Intell. Transp. Syst., vol. 19, no. 7, pp. 2204–2220, aadhar using blockchain technology,” in Proc. 2nd Int. Conf. Innov.
Jul. 2018. Mechanisms Ind. Appl., 2020, pp. 71–75.
[71] Y. Pu, T. Xiang, C. Hu, A. Alrawais, and H. Yan, “An efficient [93] R. Bosri, A. R. Uzzal, A. Al Omar, A. T. Hasan, and M. Z. A. Bhuiyan,
blockchain-based privacy preserving scheme for vehicular social net- “Towards a privacy-preserving voting system through blockchain tech-
works,” Inf. Sci., vol. 540, pp. 308–324, 2020. nologies,” in Proc. IEEE Intl. Conf. Dependable, Autonomic Secure
[72] H. Shen, J. Zhou, Z. Cao, X. Dong, and K.-K. R. Choo, “Blockchain- Comput., Intl. Conf. Pervasive Intell. Comput., Intl. Conf. Cloud Big
based lightweight certificate authority for efficient privacy-preserving Data Comput., Intl. Conf. Cyber Sci. Technol. Congr., 2019, pp. 602–
location-based service in vehicular social networks,” IEEE Internet 608.
Things J., vol. 7, no. 7, pp. 6610–6622, 2020. [94] B. Shahzad and J. Crowcroft, “Trustworthy electronic voting using ad-
[73] J. Hu, D. He, Q. Zhao, and K.-K. R. Choo, “Parking management: A justed blockchain technology,” IEEE Access, vol. 7, pp. 24477–24488,
blockchain-based privacy-preserving system,” IEEE Consum. Electron. 2019.
Mag., vol. 8, no. 4, pp. 45–49, Jul. 2019. [95] A. Meola, “Smart farming in 2020: How IoT sensors are creating a
[74] W. Al Amiri, M. Baza, K. Banawan, M. Mahmoud, W. Alasmary, and more efficient precision agriculture industry,” Bus. Insider, 2020, Ac-
K. Akkaya, “Privacy-preserving smart parking system using blockchain cessed: Dec. 4, 2020. [Online]. Available: https://www.businessinsider.
and private information retrieval,” in Proc. Int. Conf. Smart Appl., Com- com/smart-farming-iot-agriculture?r=AU&IR=T
mun. Netw. (SmartNets), 2019, pp. 1–6. [96] L. Huning, J. Bauer, and N. Aschenbruck, “A privacy preserving mobile
[75] M. Baza, M. Mahmoud, G. Srivastava, W. Alasmary, and M. Younis, crowdsensing architecture for a smart farming application,” in Proc. 1st
“A light blockchain-powered privacy-preserving organization scheme ACM Workshop Mobile Crowdsensing Syst. Appl., 2017, pp. 62–67.
for ride sharing services,” in Proc. IEEE 91st Veh. Technol. Conf.
(VTC2020-Spring), 2020, pp. 1–6. QUANG NHAT TRAN received the B.S. degree
[76] M. Li, L. Zhu, and X. Lin, “Coride: A privacy-preserving collaborative- majoring in computer science from New Mexico
ride hailing service using blockchain-assisted vehicular fog computing,” Highlands University, Las Vegas, NM, USA, in
in Proc. Int. Conf. Secur. Privacy Commun. Syst. Berlin, Germany: 2012 and the master’s degree in 2017 from the
Springer, 2019, pp. 408–422. University of New South Wales, Sydney, NSW,
[77] M. Kim, B. Hilton, Z. Burks, and J. Reyes, “Integrating blockchain, Australia, where he is currently working toward the
smart contract-tokens, and iot to design a food traceability solution,” in Ph.D. degree. In 2013, he was a Lecturer with the
Proc. IEEE 9th Annu. Inf. Technol., Electron. Mobile Commun. Conf., Department of Computer Science and Technology,
2018, pp. 335–340. Posts and Telecommunication Institute of Technol-
[78] J. Lin, Z. Shen, A. Zhang, and Y. Chai, “Blockchain and iot based food ogy, Hanoi, Vietnam. His research interests include
traceability for smart agriculture,” in Proc. 3rd Int. Conf. Crowd Sci. computer security, biometric template protection,
Eng., 2018, pp. 1–6. biometric cryptography, and blockchain technology.

VOLUME 2, 2021 83
TRAN ET AL.: SURVEY ON PRIVACY-PRESERVING BLOCKCHAIN SYSTEMS (PPBS) AND A NOVEL PPBS-BASED FRAMEWORK FOR SMART AGRICULTURE

BENJAMIN P. TURNBULL is currently a Se- KATERINA KORMUSHEVA received the Ph.D.

nior Lecturer with the Australian Centre for Cy- degree in marketing management from Australian
ber Security, University of New South Wales, Syd- National University, Canberra, ACT, Australia. She
ney, NSW, Australia. His research interests include is a Certified Practicing Marketer by the Australian
cyber-resilience, cyber-kinetic impact analysis, and Institute of Marketing and a Professional Certi-
novel methods for network analysis. He was with fied Marketer by the American Marketing Associ-
the Australian Defence Force. ation. She has published in top service and market-
ing publications such as The Journal of Services
Marketing.

HAO-TIAN WU received the B.Eng. and M.Eng.

degrees from the Harbin Institute of Technol-
ogy, Harbin, China, in 2002 and 2004, respec-
tively, and the Ph.D. degree in computer science
from Hong Kong Baptist University, Hong Kong, JIANKUN HU (Senior Member, IEEE) received
in 2007. He is currently an Associate Profes- the B.E. degree from Hunan University, Chang-
sor with the School of Computer Science and sha, China, in 1983, the Ph.D. degree in control
Engineering, South China University of Technol- engineering from the Harbin Institute of Technol-
ogy, Guangzhou, China. His research interests ogy, Harbin, China, in 1993 and the Masters by
include image enhancement, homomorphic en- Research in computer science and software engi-
cryption, privacy preservation, and reversible data neering from Monash University, Melbourne, VIC,
hiding. He has published in top publications such as the IEEE TRANSACTIONS Australia, in 2000. He is currently a Full Profes-
ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY and the IEEE SIGNAL sor with the School of Engineering and Informa-
PROCESSING LETTERS. tion Technology, University of New South Wales,
Canberra, Australia. He was with Ruhr University,
Bochum, Germany, on the prestigious German Alexander von Humboldt
Fellowship 1995-1996 and from 1998 to 1999, a Research Fellow with Mel-
bourne University, Melbourne, VIC, Australia. He has authored or coauthored
many papers in his field of research in high quality conferences and journals
including the IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE
A. J. S. DE SILVA received the Ph.D. degree INTELLIGENCE, which include cyber security, including Image Processing,
in computer science from the Australian Defence Forensics and machine learning. He was on the Editorial Board of seven
Force Academy (ADFA) and University of New international journals including the top publication IEEE TRANSACTIONS ON
South Wales. He is the Group Managing Director INFORMATION FORENSICS AND SECURITY and was the Security Symposium
of TelSoft Pty Ltd., and has implemented Telecom- Chair of the IEEE flagship conferences of the IEEE ICC and the IEEE
munications projects in more than 40 counties. He Globecom. He has obtained nine ARC (Australian Research Council) grants
is the Founding Director of Smartfarmnet Aus- and has served at the prestigious Panel of Mathematics, Information and
tralia. Computing Sciences (MIC), ARC ERA (The Excellence in Research for
Australia) Evaluation Committee.

84 VOLUME 2, 2021