1

Blockchain for Internet of Things: A Survey

Hong-Ning Dai, Senior Member, IEEE, Zibin Zheng, Senior Member, IEEE, Yan Zhang, Senior Member, IEEE

Abstract— Internet of Things (IoT) is reshaping the incumbent transaction-management systems where the centralized agency
industry to smart industry featured with data-driven decision- needs to validate the transaction, blockchains can achieve the
making. However, intrinsic features of IoT result in a number of decentralized validation of transactions, thereby greatly saving
challenges such as decentralization, poor interoperability, privacy
and security vulnerabilities. Blockchain technology brings the the cost and mitigating the performance bottleneck at the cen-
opportunities in addressing the challenges of IoT. In this paper, tral agency. Moreover, each transaction saved in blockchains
we investigate the integration of blockchain technology with IoT. is essentially immutable since each node in the network keeps
We name such synthesis of blockchain and IoT as Blockchain of all the committed transactions in the blockchain. Meanwhile,
Things (BCoT). This paper presents an in-depth survey of BCoT crytographic mechanisms (such as asymmetric encryption al-
and discusses the insights of this new paradigm. In particular,
we first briefly introduce IoT and discuss the challenges of gorithms, digital signature and hash functions) guarantee the
IoT. Then we give an overview of blockchain technology. We integrity of data blocks in the blockchains. Therefore, the
next concentrate on introducing the convergence of blockchain blockchains can ensure non-repudiation of transactions. In
and IoT and presenting the proposal of BCoT architecture. We addition, each transaction in blockchains is traceable to every
further discuss the issues about using blockchain for 5G beyond user with the attached historic timestamp.
in IoT as well as industrial applications of BCoT. Finally, we
outline the open research directions in this promising area. Blockchain is essentially a perfect complement to IoT with
the improved interoperability, privacy, security, reliability and
Index Terms— Blockchain; Internet of Things; Smart Con-
tract; Industrial Applications scalability. In this paper, we investigate a new paradigm of
integrating blockchain with IoT. We name such synthesis
of blockchain and IoT as Blockchain of Things (BCoT). In
I. I NTRODUCTION particular, BCoT has the following merits:
The recent advances in information and communication
• Interoperability across IoT devices, IoT systems and
technology (ICT) have promoted the evolution of conventional
industrial sectors, where the interoperability is the ability
computer-aided industry to smart industry featured with data-
of interacting with physical systems and exchanging
driven decision making [1]. During this paradigm shift, In-
information between IoT systems. It can be achieved
ternet of Things (IoT) plays an important role of connect-
through the blockchain-composite layer built on top of an
ing the physical industrial environment to the cyberspace of
overlay peer-to-peer (P2P) network with uniform access
computing systems consequently forming a Cyber-Physical
across different IoT systems.
System (CPS). IoT can support a wide diversity of industrial
• Traceability of IoT data, where the traceability is the
applications such as manufacturing, logistics, food industry
capability of tracing and verifying the spatial and tempo-
and utilities. IoT aims to improve operation efficiency and
ral information of a data block saved in the blockchain.
production throughput, reduce the machine downtime and
Each data block saved in a blockchain is attached with a
enhance product quality. In particular, IoT has the following
historic timestamp consequently assuring the data trace-
features: 1) decentralization of IoT systems, 2) diversity of
ability.
IoT devices and systems, 3) heterogeneity of IoT data and
• Reliability of IoT data is the quality of IoT data being
4) network complexity. All of them result in the challenges
trustworthy. It can be ensured by the integrity enforced by
including heterogeneity of IoT system, poor interoperability,
crytographic mechanisms including asymmetric encryp-
resource constraints of IoT devices, privacy and security
tion algorithms, hash functions and digital signature, all
vulnerabilities.
of which are inherent in blockchains.
The appearance of blockchain technologies brings the op-
• Autonomic interactions of IoT system refer to the capabil-
portunities in overcoming the above challenges of IoT. A
ity of IoT systems interacting with each other without the
blockchain is essentially a distributed ledger spreading over
intervention of a trusted third party. This autonomy can be
the whole distributed system. With the decentralized consen-
achieved by smart contracts enabled by blockchains. In
sus, blockchains can enable a transaction to occur and be
particular, contract clauses embedded in smart contracts
validated in a mutually-distrusted distributed system without
will be executed automatically when a certain condition
the intervention of the trusted third party. Unlike incumbent
is satisfied (e.g., the user breaching the contract will be
Corresponding authors: Zibin Zheng and Yan Zhang. punished with a fine automatically).
H.-N. Dai is with Faculty of Information Technology, Macau University of
Science and Technology, Macau (email: hndai@ieee.org). Though BCoT can benefit IoT, there are also a number of
Z. Zheng is with School of Data and Computer Science, Sun Yat-sen challenges to be addressed before the potentials of BCoT can
University, China (email: zhzibin@mail.sysu.edu.cn). be fully unleashed. Therefore, this paper aims to present an in-
Y. Zhang is with Department of Informatics, University of Oslo, Norway.
He is also with Simula Metropolitan Center for Digital Engineering, Norway depth survey on the state-of-the-art advances, challenges and
(email: yanzhang@ieee.org). open research issues in BCoT.
2

A. Comparison between this paper and existing surveys V discusses the research issues about blockchain for 5G-
There are several published papers discussing the conver- beyond networks. Section VI next summarizes the applications
gence of blockchain with IoT. For example, the work of of BCoT. Open research issues are discussed in Section VII.
[2] presents a smart home application of using blockchains Finally, the paper is concluded in Section VIII.
for IoT. Zhang and Wen [3] proposed a business model to
support P2P trading based on smart contracts and blockchains. II. I NTERNET OF T HINGS
However, these studies are too specific to a certain scenario In this section, we briefly introduce Internet of Things (IoT)
of incorporating blockchain with IoT (e.g., a smart home in Section II-A and summarize the challenges of IoT in Section
application). II-B.
Recently, several surveys on the convergence of blockchain
with IoT have been published. In particular, [4] gives a
systematic literature review on blockchain for IoT with the A. Introduction to Internet of Things
categorization of a number of use cases. The work of [5] Today’s industry is experiencing a paradigm shift from
presents a survey on IoT security and investigates the po- conventional computer-aided industry to smart industry driven
tentials of blockchain technologies as the solutions. Reyna by recently advances in Internet of Things (IoT) and Big
et al. [6] investigated the possibility and research issues of Data Analytics (BDA). During this evolution, IoT plays a
integrating blockchain with IoT. The work of [7] presents a critical role of bridging the gap between the physical industrial
review on integrating blockchain with IoT in the application environment and the cyberspace of computing systems while
aspect. Ref. [8] attempted to give a comprehensive survey BDA can help to extract hidden values from massive IoT data
on application of blockchain in IoT. The work of [9] gives so as to make intelligent decisions.
a categorization of applications of blockchain for IoT. IoT is essentially a network of smart objects (i.e., things)
However, most of the existing surveys suffer from the fol- with provision of various industrial services. A typical IoT
lowing limitations: 1) there is no general architecture proposed system consists of the following layered sub-systems (from
for BCoT; 2) there is no study explicitly discussing blockchain bottom to up) as shown in Fig. 1:
for 5G beyond networks for IoT (however, this topic is of great • Perception Layer: There is a wide diversity of IoT devices
importance for the development of IoT); 3) other important including sensors, actuators, controllers, bar code/Quick
issues like life cycle of smart contracts are missing in most of Response Code (QR Code) tags, RFID tags, smart meters
the existing surveys. and other wireless/wired devices. These devices can sense
and collect data from the physical environment. Mean-
B. Contributions while, some of them (like actuators and controllers) can
In view of prior work, we aim to (i) provide a concep- make actions on the environment.
• Communication Layer: Various wireless/wired devices
tual introduction on IoT and blockchain technologies, (ii)
present in-depth analysis on the potentials of incorporating such as sensors, RFIDs, actuators, controllers and other
blockchains into IoT and (iii) give insightful discussions of tags can then connect with IoT gateways, WiFi Access
technical challenges enabling BCoT. In summary, the main Points (APs), small base stations (BS) and macro BS
contributions of this paper are highlighted as follows: to form an industrial network. The network connection
is enabled by a diverse of communication protocols
1) A brief introduction on IoT is first given and then
such as Bluetooth, Near Field Communications (NFC),
accompanied by a summary of key characteristics of IoT.
Low-power Wireless Personal Area Networks (6LoW-
Meanwhile, research challenges of IoT are outlined.
PAN), Wireless Highway Addressable Remote Trans-
2) An overview of key blockchain technologies is
ducer (WirelessHART) [10], Low Power Wide Area Net-
then given with a summary of key characteristics
works (LPWAN) technologies including Sigfox, LoRa,
of blockchains and a taxonomy of the incumbent
Narrowband IoT (NB-IoT) and industrial Ethernet [11].
blockchain systems.
• Industrial Applications: IoT can be widely used to sup-
3) The core part of this paper is focused on the convergence
port a number of industrial applications. The typical in-
of blockchain and IoT. In this respect, the opportunities
dustrial applications include manufacturing, supply chain,
of integrating blockchain with IoT are first discussed. An
food industry, smart grid, health care and internet of
architecture of BCoT is then proposed and illustrated.
vehicles.
4) The 5G-beyond networks play an important role in con-
structing the infrastructure for BCoT. Research issues
about blockchain for 5G-beyond networks in IoT are B. Challenges of Internet of Things
also discussed. In this paper, we mainly focus on Industrial IoT. We denote
5) Furthermore, this paper summarizes the applications of Industrial IoT by IoT thereafter without loss of generality.
BCoT and outlines the open research issues in BCoT. The IoT ensures the connection of various things (smart
The remainder of the paper is organized as follows. Section objects) mounted with various electronic or mechanic sensors,
II first presents an overview on IoT. Section III then gives actuators and software systems which can sense and collect
the introduction of blockchain technology. The convergence information from the physical environment and then make
of blockchain and IoT is discussed in Section IV. Section actions on the physical environment. The unique features of
 3

Industrial promise the vulnerable privacy of IoT [15].

Applications
Manufacturing Supply chain Food Industry Health care
• Security vulnerability. The decentralization and the het-
Smart grid Internet of Vehicles
erogeneity of IoT systems also result in the difficulty in
Small BS
ensuring the security of IoT while the security is ex-
Communication
layer
Macro BS
Bluetooth
tremely important for an enterprise. The typical solutions
IoT gateway WiFi AP

Small BS WiFi AP
such as authentication, authorization and communication
encryption may not be appropriate to IoT due to the dif-
Perception ficulty in implementing the security countermeasures in
layer
Sensor Meter Surveillance
camera
Robot arm QR code Bar code Panel PC Portable PCs Reader resource-constrained IoT systems. Moreover, IoT systems
are also vulnerable to malicious attacks due to the failure
Fig. 1. Internet of Things (IoT) consists of perception layer, communication of security firmware updates in time [16].
layer and industrial applications
Discussion. Some intrinsic limitations of IoT can be over-
come via recent ICT advances. For example, ambient backscat-
ter assisted communications [17] can help IoT nodes obtain
IoT pose a number of research challenges exhibiting in the
extra energy from ambience. Meanwhile, mobile edge com-
following aspects.
puting can extend the capability of IoT nodes via offloading
• Heterogeneity of IoT systems exhibits in the heteroge- the computationally-intensive tasks to edge servers [18]. More-
neous IoT devices, heterogeneous communication proto- over, the recent advances in blockchain technologies offer po-
cols and heterogeneous IoT data types (i.e., structured, tential solutions to the challenges such as poor interoperability,
semi-structured and nonstructured). The heterogeneity is privacy and security vulnerabilities. In addition, blockchain
also the root of other challenges such as interoperability, is also beneficial to improve heterogeneity of IoT systems.
privacy and security (to be explained as follows). We will discuss these opportunities brought by blockchain
• Complexity of networks. There are a number of com- to IoT in Section IV-A after giving a briefing on blockchain
munication/network protocols coexisting in IoT. Typical technologies in Section III.
network protocols include NFC, Bluetooth, 6LoWPAN,
WirelessHART, Sigfox, LoRa and NB-IoT, all of which III. B LOCKCHAIN T ECHNOLOGIES
offer different network services. For example, 6LoWPAN
In this section, we first give an overview on blockchain tech-
and WirelessHART have typically short communication
nologies in Section III-A, then summarize the key blockchain
coverage (e.g., less than 100 meters) while LPWAN
characteristics in Section III-B and present a taxonomy of
technologies can provide the coverage from 1km to 10
blockchain platforms in Section III-D.
km [12], [13].
• Poor interoperability is the capability of IoT systems
(both hardware and software) to exchange, make use of A. Overview of Blockchain Technologies
information and collaborate with each other. Due to the 1) Blockchain: A blockchain is essentially a distributed
decentralization of IoT systems and the heterogeneity ledger spreading over the whole blockchain system [19].
of IoT systems, it is challenging to exchange the data Fig. 2 shows an exemplary blockchain consisting of a num-
between different industrial sectors, strategic centers, IoT ber of consecutively-connected blocks. Each block (with the
systems. As a result, the interoperability of IoT is difficult exception of the first block) in a blockchain points to its
to be achieved. immediately-previous block (called parent block) via an in-
• Resource constraints of IoT devices. IoT devices such verse reference that is essentially the hash value of the parent
as sensors, actuators, RFID tags and smart meters suffer block. For example, block i contains the hash of block i − 1
from limited resources including computing resource, as shown in Fig. 2. The first block of a blockchain is called
storage resource and battery power. For example, there the genesis block having no parent block. In particular, a
is no battery power for passive RFID tags that can only block structure consists of the following information: 1) block
harvest the energy from RFID readers or from ambient version (indicating the validation rules to follow), 2) the hash
environment [14]. Moreover, the resource constraints also of parent block, 3) Timestamp recording the current time in
result in the vulnerability of IoT devices to malicious seconds, 4) Nonce staring from 0 and increasing for every hash
attacks. calculation, 5) the number of transactions, 6) MerkleRoot (i.e.,
• Privacy vulnerability. Privacy is to guarantee the appro- the hash value of the root of a Merkel tree with concatenating
priate usage of IoT data while there is no disclosure the hash values of all the transactions in the block) as shown
of user private information without user consent. It is in the detailed view of Fig. 2.
challenging to preserve data privacy in IoT due to the A blockchain is continuously growing with the transactions
complexity and the decentralization of IoT systems, the being executed. When a new block is generated, all the nodes
heterogeneity of IoT systems. Moreover, it becomes a in the network will participate in the block validation. A
trend to integrate IoT with cloud computing since cloud validated block will be automatically appended at the end
computing can empower IoT with extra computing and of the blockchain via the inverse reference pointing to the
storage capabilities. However, uploading the confidential parent block. In this manner, any unauthorized alterations on
IoT data to the third-party cloud servers may also com- the previously-generated block can be easily detected since the
4

Hash of block j − 1 Hash of block j

Timestamp Nonce Timestamp Nonce

MerkleRoot MerkleRoot
Merkle tree
TX 1 TX 2 TX n TX 1 TX 2 TX n
structure
MerkleRoot Block j Block j + 1
A shorter chain is deserted

hash(TX1, TX2) … hash(TXn − 1, TXn)

Hash of block 0 Hash of block i − 1 Hash of block i Hash of block m − 1

Timestamp Nonce Timestamp Nonce Timestamp Nonce Timestamp Nonce

hash(TX1) hash(TX2) … hash(TXn)
MerkleRoot MerkleRoot MerkleRoot MerkleRoot

TX 1 TX 2 TX n TX 1 TX 2 … TX n TX 1 TX 2 TX n TX 1 TX 2 TX n TX 1 TX 2 TX n

Genesis block Block i Block i + 1 Block m

Detailed view

Fig. 2. Blockchain consists of a number of consecutively-connected blocks and the detailed view represents a Merkle tree structure (where TX represents a
transaction)

hash value of the tampered block is significantly different from to this node as a compensation for solving the puzzle.
that of the unchanged block. Moreover, since the blockchain Discrepancy solution. In a distributed system, multiple
is distributed throughout the whole network, the tampering nodes may validate blocks nearly at the same time. Meanwhile,
behavior can also be easily detected by other nodes in the the network latency can somehow result in bifurcated (or
network. forked) chains at the same time. To solve the discrepancy, most
Data integrity guarantee in blockchain. Blockchains lever- of existing blockchain systems typically maintain the longest
age cryptographic techniques to guarantee data integrity. In chain as the valid chain because the longest chain implies the
particular, there are two mechanisms in blockchains to ensure most tolerant of being compromised by adversaries. If so, a
the data integrity: 1) an ordered link list structure of blocks, in shorter chain is automatically deserted (i.e., the blue dash-line
which each newly-appended block must include the hash value box as shown in Fig. 2) and the future validation work will
of the preceding block. In this manner, a falsification on any continue on the longest chain.
of the previous blocks will invalidate the subsequent blocks. Trustfulness of PoW. The trustfulness of PoW is based on
2) Merkel Tree structure, in which each block contains a root the assumption that a majority of blockchain nodes is trustful.
hash of a Merkel tree of all the transactions. Each non-leave Generally, 51% of computational capability is regarded as
node is essentially a hash value of two concatenated values of the threshold of PoW being tolerant of malicious attacks.
its two children. Therefore, a Merkel tree is typically a binary The incentive mechanisms can encourage miners to be honest
tree. In this way, any falsification on the transactions will lead against compromising. Meanwhile, solving the puzzle typi-
to a new hash value in the above layer, consequently resulting cally requires extensive computing power. The probability of
in a falsified root hash. As a result, any falsification can be solving the puzzle at a miner is often proportional to the
easily detected. computational capability and resource of a miner [22].
2) Consensus algorithms: One of the advantages of PoW schemes require extensive computation to solve the
blockchain technologies is to validate the block trustfulness in puzzle, thereby resulting in the extensive energy consumption.
a decentralized trustless environment without the necessity of Unlike PoW, PoS requires the proof of ownership to validate
the trusted third-party authority. In distributed environment, it the trustfulness of a block since the users with more cryp-
is challenging to reach a consensus on a newly-generated block tocurrencies (i.e., more stakes) are more trustful than those
as the consensus may be biased in favor of malicious nodes. with fewer cryptocurrencies. In PBFT, each node who has the
This trustfulness validation in a decentralized environment equal right to vote for the consensus will send its voting state
can be achieved by consensus algorithms. Typical consensus to other nodes. After multiple rounds of voting procedure, the
algorithms include proof of work (PoW), proof of stake (PoS) consensus reaches.
and practical byzantine fault tolerance (PBFT) [20]. We roughly categorize typical consensus algorithms into
Take PoW as an example. The creation of a newly-generated two types: 1) Probabilistic consensus algorithms and 2) De-
block is equivalent to the solution of a computationally- terministic consensus algorithms. Table I gives the taxonomy.
difficult problem. This computationally-difficult problem (aka Probabilistic consensus algorithms including PoW, PoS and
a puzzle) can nevertheless be verifiable without difficulty Delegated proof of stake (DPOS) typically first save the
[21]. Each node in the distributed peer-to-peer (P2P) network validated block to the chain and then seek the consensus of
can participate in the validation procedure. The first node all the nodes while deterministic consensus algorithms first
who solves the puzzle can append the validated block to the consent to the block and then saved the validated block to
blockchain; this node is also called a miner. It then broadcasts the chain. Moreover, probabilistic consensus algorithms often
the validation results in the whole blockchain system, conse- result in multiple bifurcate chains and the discrepancy is
quently other nodes validating and updating the new results in solved by choosing the longest chain. In contrast, deterministic
the blockchain. A small portion of bonus will then be given consensus algorithms solve the discrepancy through multiple
 5

TABLE I 1 Transaction is initiated

TAXONOMY OF TYPICAL CONSENSUS ALGORITHMS
Alice Bob
Transaction 2 The node broadcasts the
Probabilistic Consensus Deterministic Consensus transaction to the P2P network

Consensus Saving first and then con- Consenting first and then 4 The validated transaction is
procedure senting saving then appended to other
transactions to form a block

Bifurcation
Yes No TX 1 TX 2 TX n
(fork) newly added
Forming a
Arbitration Choosing the longest chain Voting to solve discrepancy block

mecha- when there are multiple through multiple Hash of block i − 1

nism forked chains communication-rounds 3 The P2P network validates Timestamp Nonce
the transcaction
Adversary MerkleRoot
< 50% computing or stakes < 1/3 voting nodes
tolerance TX 1 TX 2 TX n

Block i
High computational-
Complexity High network-complexity
complexity
Fig. 3. Working flow of blockchains
PBFT and PBFT variants,
Examples PoW, PoS, DPOS
Tendermint

rounds of communications in the overlay network. • Immutability. A blockchain consists of a consecutively-

There are many attempts to improve incumbent consensus linked chain of blocks, in which each link is essentially
algorithms, such as Ripple [23], Algorand [24], Tendermint, an inverse hash point of previous block. Any modification
proof of authority (PoA) [25], proof of elapsed time (PoET) on the previous block invalidates all the consequently-
[26]. Instead of choosing single consensus algorithm, there is generated blocks. Meanwhile, the root hash of the Merkle
a trend of integrating multiple consensus algorithms to fulfill tree saves the hash of all the committed transactions. Any
the requirements from different applications. (even tiny) changes on any transactions generates a new
3) Working flow of blockchains: We next show how a Merkle root. Therefore, any falsification can be easily
blockchain works in an example. Take a money transfer as detected. The integration of the inverse hash point and
an example as shown in Fig. 3. Alice wants to transfer an the Merkle tree can guarantee the data integrity.
amount of money to Bob. She first initiates the transaction • Non-repudiation. Recall the fact that the private key is
at a computer through her Bitcoin wallet (i.e., Step 1 ). used to put the signature to the transaction, which can
The transaction includes the information such as the sender’s then be accessible and verified by others via the cor-
wallet, the receiver’s address and the amount of money. The responding public key. Therefore, the crytographically-
transaction is essentially signed by Alice’s private key and can signed transaction cannot be denied by the transaction
be accessible and verifiable by other users via Alice’s public initiator.
key thereafter. Then the computer broadcasts the initiated • Transparency. For most of public blockchain systems
transaction to other computers (or nodes) in the P2P network (like Bitcoin and Ethereum), every user can access and
(i.e., Step 2 ). Next, a validated transaction is then appended interact with the blockchain network with an equal right.
to the end of the chain of transactions consequently forming a Moreover, every new transaction is validated and saved
new block in the blockchain once a miner successfully solves in the blockchain, consequently being available for ev-
the puzzle (i.e., Step 3 ). Finally, every node saves a replica ery user. Therefore, the blockchain data is essentially
of the updated blockchain when the validated transaction is transparent to every user who can access and verify the
appended to the blockchain (i.e., Step 4 ). committed transactions in the blockchain.
• Pseudonymity. Despite the transparency of blockchain
data, blockchain systems can preserve a certain level of
B. Key Characteristics of Blockchain
the privacy via making blockchain addresses anonymous.
In summary, blockchain technologies have the following key For example, the work of [27] presents an application
characteristics. of blockchain to preserve the privacy of personal data.
• Decentralization. In traditional transaction management However, blockchain can only preserve the privacy at a
systems, the transaction validation has been conducted certain level since blockchain addresses are essentially
through a trusted agency (e.g., a bank or government). traceable by inference [8]. For example, it is shown in
This centralization manner inevitably results in the extra [28] that the analysis of blockchain data can help to detect
cost, the performance bottleneck and the single-point fraud and illegal transactions. Therefore, blockchain can
failure (SPF) at centralized service providers. In contrast, only preserve the pseudonymity instead of full privacy.
blockchain allows the transaction being validated between • Traceability. Each transaction saved in the blockchain is
two peers without the authentication, jurisdiction or in- attached with a timestamp (recorded when the transaction
tervention done by the central agency, thereby reducing occurs). Therefore, users can easily verify and trace
the service cost, mitigating the performance bottleneck, the origins of historical data items after analyzing the
lowering the SPF risk. blockchain data with corresponding timestamps.
6

Creation Deployment Execution Completion

contracts can then be deployed to platforms on top of
Negotiation Smart contract Deployment Freezing assets Evaluation Auto-execute
Updating states
Unfreezing assets
blockchains. Contracts stored on the blockchains cannot
be modified due to the immutability of blockchains. Any
Write to
blockchain
Write to
blockchain
Write to
blockchain
emendation requires the creation of a new contract. Once
the smart contracts are deployed on blockchains, all the
parties can access the contracts through the blockchains.
Block 0 Block 1 Block i
Blockchain
Block i + 1 Block m
Moreover, digital assets of both involved parties in the
smart contract are locked via freezing the corresponding
Fig. 4. Life cycle of smart contracts consisting of four consecutive phases: digital wallets [32]. For example, the coin transfers
Creation, Deployment, Execution and Completion (either incoming or outgoing) on the wallets relevant
to the contract are blocked. Meanwhile, the parties can
be identified by their digital wallets.
C. Smart Contract 3) Execution of smart contracts. After the deployment of
Smart contracts are a great advance for blockchain tech- smart contracts, the contractual clauses have been mon-
nology [29]. In 1990s, smart contracts were proposed as a itored and evaluated. Once the contractual conditions
computerized transaction protocol that executes the contractual reach (e.g., product reception), the contractual proce-
terms of an agreement [30]. Contractual clauses that are dures (or functions) will be automatically executed. It
embedded in smart contracts will be enforced automatically is worth noting that a smart contract consisting of a
when a certain condition is satisfied (e.g., one party who number of declarative statements with logical connec-
breaches the contract will be punished automatically). tions. When a condition is triggered, the corresponding
Blockchains are enabling smart contracts. Essentially, smart statement will be automatically executed, consequently
contracts are implemented on top of blockchains. The ap- a transaction being executed and validated by miners in
proved contractual clauses are converted into executable com- the blockchains [33]. The committed transactions and
puter programs. The logical connections between contractual the updated states have been stored on the blockchains
clauses have also been preserved in the form of logical flows thereafter.
in programs (e.g., if-else-if statement). The execution of 4) Completion of smart contracts. After a smart contract
each contract statement is recorded as an immutable transac- has been executed, new states of all involved parties
tion stored in the blockchain. Smart contracts guarantee appro- are updated. Accordingly, the transactions during the
priate access control and contract enforcement. In particular, execution of the smart contracts as well as the updated
developers can assign access permission for each function in states are stored in blockchains. Meanwhile, the digital
the contract. Contract enforcement ensures that the contract assets have been transferred from one party to another
execution is deterministic. Once any conditions in a smart con- party (e.g., money transfer from the buyer to the sup-
tract are satisfied, the triggered statement will automatically plier). Consequently, digital assets of involved parties
execute the corresponding function in a predictable manner. have been unlocked. The smart contract has completed
For example, Alice and Bob agree on the penalty of violating the whole life cycle.
the contract. If Bob breaches the contract, the corresponding It is worth mentioning that during deployment, execution
penalty (as specified in the contract) will be automatically paid and completion of a smart contract, a sequence of transactions
from Bob’s deposit. has been executed (each corresponding to a statement in the
The whole life cycle of smart contracts consists of four smart contract) and stored in the blockchain. Therefore, all the
consecutive phases as illustrated in Fig. 4: three phases need to write data to the blockchain as shown in
1) Creation of smart contracts. Several involved parties Fig. 4.
first negotiate on the obligations, rights and prohibitions
on contracts. After multiple rounds of discussions and D. Taxonomy of Blockchain Systems
negotiations, an agreement can reach. Lawyers or coun- We classify blockchain systems into three types: 1) public
selors will help parties to draft an initial contractual blockchains, 2) private blockchains and 3) consortium (or
agreement. Software engineers then convert this agree- community) blockchains [39]. Most digital currencies such as
ment written in natural languages into a smart contract BTC (i.e., the ticker symbol of Bitcoin cryptocurrency) and
written in computer languages including declarative lan- ETH (i.e., the ticker symbol of Ethereum cryptocurrency) are
guage and logic-based rule language [31]. Similar to implemented on public blockchains, thereby being accessible
the development of computer software, the procedure by anyone in the P2P network. Differently, private blockchains
of the smart contract conversion is composed of design, can be managed or controlled by a single organization while
implementation and validation (i.e., testing). It is worth consortium blockchains sit in limbo between public and private
mentioning that the creation of smart contracts is an blockchains. Table II presents a comparison of three types of
iterative process involving with multiple rounds of ne- blockchains.
gotiations and iterations. Meanwhile, it is also involved In particular, we summary the comparison among public,
with multiple parties, such as stakeholders, lawyers and private and consortium blockchains in the following aspects.
software engineers. • Key characteristics. Public blockchains are fully-
2) Deployment of smart contracts. The validated smart decentralized while private and consortium blockchains
 7

TABLE II
tion. Private blockchains can easily achieve the consensus
C OMPARISONS OF B LOCKCHAIN SYSTEMS
among the authenticated users. Typical consensus algo-
Public Private Consortium rithms used for private blockchains include PBFT, PoA
and PoET. Moreover, consortium blockchains are a hybrid
Partially Decentral- type of public blockchains and private blockchains. In
Decentralization Decentralized Centralized
ized
particular, Ripple [23] is a variant of PBFT typically used
Partially
Immutability Immutable Alterable
Immutable for consortium blockchains.
• Exemplary platforms. Bitcoin [34] and Ethereum [35]
Non- Non-
Refusable Partially Refusable are two typical public blockchain platforms, which are
repudiation refusable
Partially Transpar- mainly used for digital currency. With regard to pri-
Transparency Transparent Opaque vate blockchains, GemOS [36] is a private blockchain
ent
Traceability Traceable Traceable Partially Traceable platform for healthcare and supply chain. In addition,
MultiChain [37] is an open source platform granting the
Scalability Poor Superior Good
implementation of private blockchains. As for consortium
Flexibility Poor Superior Good blockchains, Hyperledger [38] is developing business
Permission Permissionless Permissioned Permissioned consortium blockchain frameworks. Moreover, Ethereum
Consensus PoW, PoS Ripple PBFT, PoA, PoET also provides tools for building consortium blockchains
GemOS [40].
Bitcoin [34],
[36], Hyperledger [38]
Examples Ethereum
Multichain Ethereum [35] IV. C ONVERGENCE OF B LOCKCHAIN AND I OT
[35]
[37]
In this section, we first discuss the opportunities of integrat-
ing blockchain with IoT in Section IV-A. We then present the
architecture of the integration of blockchain and IoT (namely
are partially decentralized or fully controlled by a single BCoT) in Section IV-B. We next discuss the deployment issues
group or multiple groups. Moreover, it is nearly impossi- on BCoT in Section IV-C.
ble to tamper transactions in public blockchains as every
node keeps a replica of the blockchain (containing all the
transactions) while the dominant organization or multiple A. Opportunities of integrating blockchain with IoT
parties of consortium and private blockchains can modify As summarized in Section II-B, IoT systems are facing
the blockchain. Similarly, public blockchains can fully many challenges such as heterogeneity of IoT systems, poor
ensure the non-repudiation, transparency and traceability interoperability, resource constraints of IoT devices, privacy
of transactions while private and consortium blockchains and security vulnerabilities. Blockchain technologies can com-
cannot or can only partially ensure these properties. plement IoT systems with the enhanced interoperability and
• Scalability. Although public blockchains can guaran- the improved privacy and security. Moreover, blockchain can
tee the decentralization, immutability, transparency, non- also enhance the reliability and scalability of IoT systems [6].
repudiation and traceability, the merits are obtained in the In short, we name such integration of blockchain with IoT as
cost of low transaction-validation rate, high latency and BCoT. BCoT has the following potential benefits in contrast
extra storage space consumption, consequently limiting to incumbent IoT systems.
the scalability of public blockchains. Compared with • Enhanced interoperability of IoT systems. Blockchain
public blockchains, private and consortium blockchains can essentially improve the interoperability of IoT
have a better scalability since blockchains are fully con- systems via transforming and storing IoT data into
trolled by a single group or multiple organizations and blockchains. During this procedure, heterogeneous types
the consensus can be easily reached. of IoT data are converted, processed, extracted, com-
• Flexibility. Similarly, public blockchains have the less pressed and finally stored in blockchains. Moreover, the
flexibility than private and consortium blockchains since interoperability also exhibits in easily passing through
configurations of private and consortium blockchains are different types of fragmented networks since blockchains
more adjustable. are established on top of the P2P overlay network that
• Permission. Permission refers to consent or authorization supports universal internet access.
to access the blockchains. In public blockchains, public • Improved security of IoT systems. On one hand, IoT
participation is allowed, thereby being permissionless. data can be secured by blockchains since they are
However, private and consortium blockchains can allow stored as blockchain transactions which are encrypted and
one or more users to access and interact with blockchains digitally-signed by cryptographic keys (e.g., elliptic curve
with different permission levels. For example, some users digital signature algorithm [41]). Moreover, the integra-
can only read the blockchain data while others can either tion of IoT systems with blockchain technologies (like
read or initiate transactions. smart contracts) can help to improve the security of IoT
• Consensus. Public blockchains usually use PoW and PoS systems by automatically-updating IoT device firmwares
as the consensus algorithms, which are Byzantine-failure to remedy vulnerable breaches thereby improving the
tolerant while resulting in extensive resource consump- system security [42].
8

Industrial
Applications Data block
Manufacturing Supply chain Food Industry Health care Chain structure
Smart grid Internet of Vehicles
Blockchain as a service Peer Merkle tree
(BaaS)
Service sub-layer Smart contracts
Hash function
Currency issue/ Reward Transaction
distribution mechanism Incentive sub-layer mechanism cost (fee) Cryptographic
Blockchain- algorithms
PoW PoS Consensus sub-layer PBFT DPOS
composite layer Digital signature
Data store
Propagration Overlay Verification
Network sub-layer
Hash of block 0 Hash of block
Has o

protocol routing mechanism Timestamp Nonce

Timestamp Nonce Timestamp Nonce

Overlay network
Genesis block Block

Data Chain Merkle Hash Crypotograhic Digital Blockchain

Data sub-layer
block structure tree function algorithms signature sub-layer Blockchain node architecture

Communication
Small BS
Communication Macro BS layer
Bluetooth
layer IoT gateway WiFi AP
Small BS
Macro BS WiFi AP
IoT gateway
Small BS WiFi AP
Perception
Perception layer Meter
Sensor
layer Sensor
Sensor Meter Surveillance Robot arm QR code Bar code Panel PC Portable PCs Reader Meter Reader
camera Robot arm

(a) Blockchain-composite layer (b) P2P overlay network and blockchain node architecture

Fig. 5. Overview of BCoT architecture

• Traceability and Reliability of IoT data. Blockchain data data with digital signature via asymmetric cryptographic
can be identified and verified anywhere and anytime. algorithms and hash functions. These consecutively-
Meanwhile, all the historical transactions stored in the connected data blocks then form the blockchain after
blockchains are traceable. For example, the work of [43] the distributed validation. Different blockchain platforms
has developed a blockchain-based product traceability may choose different cryptographic algorithms and hash
system, which provide suppliers and retailers with trace- functions. For example, Bitcoin blockchain chooses
able services. In this manner, the quality and originality of SHA-256 as the hash function and elliptic curve digital
the products can be inspected and verified. Moreover, the signature algorithm (ECDSA) as the signature algorithm.
immutability of blockchains also assures the reliability of 2) Network sub-layer is essentially an overlay P2P net-
IoT data since it is nearly impossible to alter or falsify work running on top of the communication layer. The
any transactions stored in blockchains. overlay network consists of either virtual or physical
• Autonomic interactions of IoT systems. Blockchain tech- links connecting nodes in the underlying communication
nologies can grant IoT devices or subsystems to interact networks (i.e., wired/wireless communication networks).
with each other automatically. For example, the work One node only simply broadcasts the block of transac-
of [44] proposes Distributed autonomous Corporations tions to its connected peers. Once receiving the block
(DACs) to automate transactions, in which there are no of transactions, other peers will verify it locally. If it
traditional roles like governments or companies involved is valid, the block will be further propagated to other
with the payment. Being implemented by smart contracts, nodes through the overlay network.
DACs can work automatically without human interven- 3) Consensus sub-layer is mainly involved with the dis-
tion consequently saving the cost. tributed consensus for the trustfulness of a block. The
consensus can be achieved by various consensus algo-
B. Architecture of Blockchain of Things rithms like PoW, PoS, PBFT and DPOS (as explained
We propose the architecture of BCoT as shown in Fig. 5. in Section III-A.2). It is worth mentioning that block
In this architecture, the blockchain-composite layer plays as propagation mechanisms (such as relay network propa-
a middleware between IoT and industrial applications. This gration and advertisement-based propagation [21]) are
design has two merits: 1) offering an abstraction from the the prerequisite for the distributed consensus protocols.
lower layers in IoT and 2) providing users with blockchain- 4) Incentive sub-layer is responsible for the following
based services. In particular, the blockchain-composite layer tasks: 1) digital currency issuing, 2) digital currency dis-
hides the heterogeneity of lower layers (like perception layer tribution, 3) designing reward mechanism (especially for
and communication layer in IoT). On the other hand, the miners), 4) handling transaction cost, etc. In particular,
blockchain-composite layer offers a number of blockchain- it is important to design appropriate monetary policy of
based services, which are essentially application programming digital currency (i.e., money creation and distribution),
interfaces (APIs) to support various industrial applications. As distribute rewards to participants who contribute to dis-
a result, the difficulty of developing industrial applications can tributed consensus (i.e., mining).
also be lowered down due to the abstraction achieved by the 5) Service sub-layer provides users with blockchain-based
blockchain-composite layer. services for various industrial sectors include manu-
In particular, the blockchain-composite layer consists of 5 facturing, logistics, supply chains, food industry and
sub-layers as shown in Fig. 5(a) (from bottom to up): utilities. The blockchain as a service (BaaS) can be
1) Data sub-layer collects the IoT data from the lower lay- achieved by smart contracts, which can be automatically
ers (e.g., perception layer) and wraps up the encrypted triggered when a special event occurs. For example,
 9

Wired link
Wireless link
MEC server
Cloud server Blockchain
Cloud server

Meter MBS IoT gateway

Blockchain SBS
Blockchain Blockchain
D2
617d48c9 Dl 617d48c9
ink
Partial (hash) of 617d48c9 Partial (hash) of 617d48c9

blockchain Partial (hash) of blockchain 617d48c9

sensor Partial (hash) of
blockchain Partial (hash) of blockchain
blockchain

Fig. 6. Deployment scenario of BCoT

a payment contract is automatically executed when a only save the particial blockchain data. In addition to the
product is well received by a consumer. deployment of BCoT, there are also several possible interaction
It is worth mentioning that the network sub-layer that is manners between IoT and blockchain [8]: (i) direct interaction
established on top of the communication layer is the abstrac- between IoT and blockchain, in which IoT devices can directly
tion of underneath communication networks, consequently access blockchain data saved at edge servers co-located with
offering a universal network access across different networks IoT gateways, Macro Base Stations (MBS) or Small BS; (ii)
as shown in Fig. 5(b). Fig. 5(b) also shows the architecture of direct interaction between IoT nodes, in which IoT nodes
a blockchain node, which essentially includes blockchain data can directly exchange/access partial blockchain data via D2D
and other elements in the data sub-layer. links; (iii) hybrid interaction of cloud and edge servers with
IoT devices, in which IoT devices can interact with blockchain
data through edge/cloud servers.
C. Deployment of BCoT There are several initiatives addressing the configuration
The realistic deployment of BCoT is of great importance. and initialization of blockchain at edge servers or at IoT
However, due to the constraints of IoT devices, it is challeng- devices. For example, Raspnode2 is a project mainly for
ing to store the whole blockchain at IoT devices. In particular, installing Bitcoin and other blockchains at Raspberry Pi micro
there are two modes to store the blockchain data [6]: i) full computers. EthArmbian3 offers the customized Ubuntu Linux
storage, in which the entire blockchain is stored, ii) partial image for ARM devices, each of which can serve as an
storage, in which only a subset of data blocks are stored Ethereum node. Despite these initiatives, most of IoT devices
locally. Accordingly, we name the nodes with full storage are still lightweight nodes due to the limited storage.
of blockchain data as full nodes and the nodes with partial
storage of blockchain data as lightweight nodes. In practice, V. B LOCKCHAIN FOR 5G B EYOND IN I OT
a full node can be a cloud server or an edge server with
Although blockchain technology is promising to IoT, there
adequate computing resources since it requires a large storage
are still many research issues to be addressed before the
space to save the entire blockchain (e.g., the whole Bitcoin
integration of blockchain with IoT, especially for the next-
blockchain occupies nearly 185 GB at the end of September
generation networks (i.e., 5G-beyond or 6G networks), which
2018 according to the statistic report1 ) and strong computing
play a critical role in constructing the infrastructure for
capability of solving consensus puzzles (i.e., mining). On the
blockchains. Fig. 7 illustrates the potentials brought by
other hand, resource-constrained IoT devices (e.g., sensors,
blockchain to 5G-beyond networks in the perspectives from
IoT objects) can be lightweight nodes that can validate the
communications, network management and computing man-
trustfulness of a transaction without downloading or saving
agement. We explain them in details as follows.
the whole blockchain (i.e., only saving partial blockchain
data such as hash values). It is worth mentioning that the
lightweight nodes highly rely on the full nodes. A. Blockchain for communications
Fig. 6 presents a possible deployment scenario of BCoT, The growing demands of mobile data traffic are driving the
in which cloud servers and edge servers may store the whole more efficient resource management in the fifth generation
blockchain (or partial blockchain) data while IoT devices may (5G) communication systems. For example, radio spectrum
1 https://www.statista.com/statistics/647523/worldwide-bitcoin-blockchain- 2 http://raspnode.com/

size/ 3 http://raspnode.com/
10

Computing Management
Computing resource Storage (cache)
Management management

Cloud server MEC Server Storage Storage

Network Management
SDN Control Network Network Performance
Software Virtualization Slicing control

Network device
Network device
Network device Network controller
Commmunication Management
Spectrum Service
Management Management

Macro BS Small BS
Block 0 Block i Block i + 1 Block m
IoT gateway Blockchain
WiFi AP
Small BS

Fig. 7. Blockchain for 5G Beyond Networks in IoT

is one of the most important resources [45]. Radio spectrum [54]. However, it is shown in [55] that the centralization
management typically includes spectrum auction and spectrum of SDN can also result in the single-point-of-failure. More-
sharing. It is shown in the latest speech [46] given by Federal over, incumbent SDN devices (such as gateways) are also
Communications Commission (FCC) commissioner J. Rosen- incapable of conducting computational-intensive analysis on
worcel that blockchain technology could be used to achieve data traffic. The integration of blockchain technology with
the dynamic and secure spectrum management in 5G and SDN can overcome the disadvantages of SDN. For example,
5G beyond (aka 6G) communication systems [47], [48]. The the work of [56] proposes a secure blockchain-based SDN
benefits of using blockchains for 5G-beyond networks lie in framework for IoT. In particular, a blockchain-based scheme
the secure and traceable transaction-management without the has been developed to update the flow rule table in a secure
necessity of a central intermediary, consequently saving the way without the necessity of the intermediary. In addition,
management cost. Ref. [49] gives several use cases to illustrate blockchain can also help to secure the network management
that using blockchain technology can benefit radio spectrum of network function visualization (NFV). In particular, it is
sharing in terms of trustfulness, consensus and cost reduction. shown in [57] that the integration of blockchain with NFV can
Moreover, Kotobi and Bilen [50] put forth a blockchain-based ensure that the configuration of NFV is immutable, auditable,
protocol to secure spectrum sharing between primary users and non-repudiable, consistent and anonymous. A prototype of the
cognitive users in wireless communication systems. In addi- proposed architecture was also developed and implemented in
tion, blockchain may potentially help to share link conditions this work.
to multiple IoT nodes with privacy preservation consequently In addition to SDN and NFV, the appearance of network
improving spectral efficiency via traffic optimization [51]. slicing technologies [58] brings the agility and flexibility
In addition to the radio spectrum management, blockchains of networks to support different functional and performance
also have the potentials to provide users with the improved requirements. As mentioned in Section IV, different industrial
mobile services. For example, 5G networks typically consist of sectors have diverse application demands on blockchains. For
a number of fragmented heterogeneous networks. Blockchains example, a single blockchain is typically used in digital-
that are built on top of the network layer can help to integrate currency like applications while an enterprise may main-
different networks with the provision of seamless access tain several blockchains to serve for different purposes. In
between different networks. Moreover, smart contracts can particular, four isolated blockchains are dedicate to Enter-
automate the procedure of provisions and agreements between prise resource planning (ERP), Product Lifecycle Manage-
network operators and subscribers while operational cost can ment (PLM), Manufacturing execution systems (MES) and
be greatly saved [52]. The work of [53] also shows that a Customer Relationship Management (CRM), respectively [59].
blockchain-based system can help operating nodes to improve Network slicing can essentially offer a solution to the diverse
their operational and service capabilities. In the future, the demands of blockchain applications in mobile edge computing.
synthesis of blockchains and big data analytics can help For example, each of network instances can be created for the
service providers to extract valuable insights from transactions provision of a specific blockchain service on top of network
of subscribers and offer the better services for users. slicing and network visualization. However, it is necessary to
optimize and allocate both network and computing resources
B. Blockchain for network management to fulfill the diverse demands in the composite environment
Recently, software defined networking (SDN) technology of mobile edge computing and cloud computing. Moreover,
can bestow the flexibility and scalability for distributed IoT the integration of blockchain and network slicing technologies
 11

Smart manufacturing Supply chain

can also support the reliable content sharing in content-centric
networks (CCNs) [60] and privacy preservation in data sharing Raw material Retailer

in 5G networks [61]. Factory Distribution

C. Blockchain for computing management

Due to the resource constraints of IoT devices, massive IoT
Smart grid Food industry
data has been typically uploaded to remote cloud servers for
further processing. However, the pure cloud-based computing
Power plant Smart meter
paradigm also causes the network traffic bottlenecks, long la- sensor

Blockchain of Things
tency, context unawareness and privacy exposure [62], thereby Transmission Renewable energy

limiting the scalability of IoT. Recently, Mobile Edge Com-

puting (MEC) [63] is becoming a crucial complement to cloud
computing by offloading computing tasks from distant cloud Health care Internet of Vehicles
and UAV
servers to MEC servers typically installed at IoT gateways,
WiFi APs, Macro BS and Small BS, which are close to users. RSU UAV

In this manner, the context-aware, latency-critical and less-

computing-intensive tasks can be migrated from remote cloud
servers to local MEC servers, thereby improving the response, Fig. 8. Applications of Blockchain of Things
privacy-preservation and context-awareness.
Blockchain technology has been applied in a variety of
fields due to its capability of establishing trust in a decen- consider a task is either done at an edge sever or at a cloud.
tralized fashion. There are still a number of issues needed In realistic application, a task can be partitioned into multiple
to be solved before MEC can be used in BCoT [64]. In sub-tasks with task dependency and those sub-tasks can be
contrast to cloud servers with strong computing capability either executed at the edge server or at the cloud server. It is
and extensive storage space, mobile edge servers usually worthwhile to investigate the task partition with consideration
have inferior capability. Moreover, mobile edge servers are of sub-task dependency in blockchains in the future.
heterogeneous in terms of computing capability, main memory,
storage space and network connection. As a result, mobile VI. A PPLICATIONS OF B LOCKCHAIN OF T HINGS
edge servers cannot accommodate the computational demands
alone. For example, a mobile edge server may not be able There is a growing trend in applying blockchain in IoT since
to solve the consensus puzzle in blockchains while a cloud blockchain technologies can help to overcome the challenges
server can serve for this goal. Therefore, it is worthwhile to of IoT. We then provide an overview of the applications of
investigate the orchestration of mobile edge computing and BCoT. It is worth mentioning that there is a wide diversity of
cloud computing for the provision of blockchain services [65]. applications of blockchains (ranging from smart manufacturing
to internet of vehicles and unmanned aerial vehicles). In this
paper, we mainly focus on the industrial applications of BCoT.
D. Orchestration of cloud and edge computing with
We roughly categorize the applications of BCoT into six types
blockchain
as shown in Fig. 8.
During the orchestration of cloud and edge computing with
blockchain, there are several challenges including computa-
tional task offloading and incentivizing resource sharing. A. Smart manufacturing
Offloading the computational tasks to edge servers can sig- The manufacturing industry is experiencing an upgrading
nificantly reduce the delay. Therefore, it is crucial to conduct from automated manufacturing to “smart manufacturing” [70].
edge-cloud interoperation [66]. Nevertheless, it can cause a Big data analytics on manufacturing data plays an important
performance bottleneck and a single-point-of-failure if all the role during this upgrading process. Massive data is generated
nodes offload their tasks to the same MEC server. The work during every phase of the product life cycle consisting of
of [67] presents an offloading method with consideration of product designing, raw material supply, manufacturing, dis-
load balancing among multiple MEC servers. Meanwhile, it is tribution, retail and after-sales service. However, the manu-
worthwhile to investigate how to incentivize both edge severs facturing data is highly fragmented, consequently leading to
and cloud servers. For example, [68] presents a contract-match the difficulty in data aggregation and data analytics. BCoT
approach to allocate computational resource and assign tasks can address the interoperability issue by interconnecting IoT
while incentivizing edge severs and cloud servers effectively. systems via P2P network and allowing data sharing across
Moreover, it is challenging to design an optimal solution to the industrial sectors. For example, several distributed blockchains
offloading tasks with consideration of spectrum, computation can be constructed to serve for different sectors and each
and energy consumption together. The work of [69] essentially blockchain is serving for a sector or more than one sector.
provides a solution to optimize the offloading energy con- BCoT can also improve the security of smart manufac-
sumption with consideration of feasible modulation schemes turing. One of major bottlenecks limiting the upgrading of
and tasks scheduling. However, most of existing studies only factories is that the IoT systems have been maintained in
12

a centralized way. For example, IoT firmware needs to be During this procedure, blockchain technologies can ensure the
upgraded regularly to remedy security breaches. However, traceability and the provenance of food industry data.
most of the firmware updates are downloaded from a central There are several proposals in this aspect. For example, the
server and then are manually installed at IoT devices. It is work of [78] proposed to use RFID and blockchain technology
expensive and in-efficient to install and upgrade the firmware to establish a supply chain platform from agriculture to food
updates in distributed IoT. The work of [42] presents an production in China. This system has demonstrated to guar-
automatic firmware upgrading solution based on smart contract antee the traceability of food supply-chain data. Meanwhile,
and blockchains. In particular, smart contracts describing the the work of [79] shows that blockchain technologies can help
firmware upgrading manners (e.g., when and where to upgrade to improve food safety via the provision of the traceable food
firmwares) are deployed across the whole industrial network. products. Moreover, it is shown in [80] that the integration of
Devices can then download and install the firmware hashes via blockchain in food supply chain can allow customers to track
smart contracts being automatically executed. As a result, the the whole process of food production. Authors also gave a
security maintenance cost can be greatly saved. In addition, a user case of using blockchain for the organic coffee industry
decentralized blockchain-based automatic production platform in Colombian. Furthermore, [81] proposes a food safety trace-
was proposed in [71] to offer a better security and privacy ability system based on the blockchain and Electronic Product
protection than conventional centralized architecture. Code (EPC) IoT tags. In particular, this system can prevent
data tampering and privacy exposure via smart contracts. A
prototype of the proposed architecture has been implemented
B. Supply chain management to demonstrate the effectiveness.
A product often consists of multiple parts provided by dif-
ferent manufacturers across countries. However, some forged D. Smart grid
(or low-quality) parts may seep into the supply chain. It is
The appearance of distributed renewable energy resources is
quite expensive to apply anti-fraud technologies in every part
reshaping the role of energy consumers from pure consumers
of a product. The integration of blockchain and IoT can solve
to prosumers who can also generate energy (e.g., from re-
this problem. In particular, every part will be associated with
newable energy resources) in addition to consuming energy
a unique ID with the creation. Meanwhile, an immutable
only [82]. Energy prosumers who have extra energy can sell
timestamp is also attached with this ID. The identification
it to other consumers. We name the energy trading between a
of every part can then be saved into a blockchain, which is
prosumer and a consumer (i.e., peers) as P2P energy trading.
tamper-resistant and traceable. For example, the work of [72]
However, it is challenging to ensure the secured and trusted
shows that the part ownership of a product can be authenticated
energy trading between two trading parties in the distributed
through a blockchain-based system. Moreover, the work of
environment.
[73] presents a traceability ontology with the integration of IoT
The appearance of blockchain technology brings the op-
and blockchain technologies based on Ethereum blockchain
portunities to ensure the secured P2P energy trading. Some
platform. The proposed framework has demonstrated to guar-
of recent studies proposed using blockchain technologies to
antee data provenance of supply chain.
tackle these challenges. For example, the work in [83] de-
On the other hand, BCoT can also be used to reduce the
veloped a secure energy trading system based on consortium
costs in after-sale services in the supply chain management.
blockchains. This system can greatly save the trading cost
The work of [74] shows a user case of a motor insurance,
without going through a central broker via the distributed
in which the settlement of claims can be automated via
consensus of blockchains. Moreover, Aitzhan and Svetinovic
smart contracts based on blockchains, thereby improving the
[84] developed a decentralized energy-trading system based on
efficiency and reducing the claim-processing time. Moreover,
blockchain technology. This system demonstrated the effec-
it is shown in [75] that integrating blockchain with IoT can
tiveness in protecting confidential energy-trading transaction
help to reduce the cost, fasten the speed and reduce the risk
in decentralized smart grid systems. Furthermore, the work
in the supply chain management. Furthermore, a blockchain-
of [85] proposed a blockchain based mechanism to provide
based Machine Learning platform [76] was proposed to secure
a secure and transparent energy demand-side management on
the data sharing among different enterprises to improve the
smart grid.
quality of customer service.

E. Health care
C. Food industry Health care becomes one of the major social-economic
BCoT can enhance the visibility of the product life cycle problems due to the aging population; it poses new chal-
especially in food industry. In particular, the traceability of lenges in traditional healthcare services because of the limited
food products is a necessity to ensure food safety. However, hospital resources. The recent advances in wearable health-
it is challenging for the incumbent IoT to guarantee the food care devices as well as BDA in health-care data bring the
traceability in the whole food supply chain [77]. For example, opportunities in promoting the remote health-care services at
a food company may be provisioned by a number of suppliers. home or at clinic. As a result, the burden of the hospital
The traceability requires digitizing the information of raw resources can be potentially released [86]. For example, senior
materials from sources to every sector of food manufacturing. citizens staying at their homes are wearing the health-care
 13

TABLE III
devices at their bodies. These wearable devices continuously
C OMPARISON OF APPLICATIONS OF BLOCKCHAIN OF THINGS
measure and collect health-care data including heart beat
rate, blood sugar and blood pressure readings. Doctors and Application Benefits
health-care teams can access health-care data at any time and
Smart manufacturing XImproving interoperability
anywhere via the health-care networks. However, assessing [42], [70], [71]
health-care data also brings privacy and security concerns. The XAutomating P2P business trading
vulnerability of health-care devices and the heterogeneity of XReducing cost for trusted third party
health-care networks pose the challenges in preserving privacy XAssuring data provenance
an ensuring security of health-care data. Supply chain
management [72]–[76] XReducing the costs in after-sale services
Incorporating blockchains into health-care networks can
XMitigating the supply chain risk
potentially overcome the challenges in privacy preservation
and security assurance of health-care data. For example, the XImproving data traceability
Food industry [77]–[81]
work of [87] shows that using blockchain technology can XEnhancing food safety
protect health-care data stored in cloud servers. Meanwhile, XSecuring energy trading
Griggs et al. [88] developed a blockchain-based system to Smart grid [82]–[85] XImproving transparency
assure the private health-care data management. In particular,
XPreserving privacy
the health-care data generated by medical sensors can be
automatically collected and transmitted to the system via exe- XAssuring security

cuting smart contracts, consequently supporting the real-time Health care [86]–[91] XPreserving privacy
patient monitoring. During the whole procedure, the privacy XVerifying authenticity
can be preserved via underneath blockchains. Moreover, the XAssuring trustworthiness of messages
work of [89] proposed a blockchain-based solution to manage IoV and UAVs
[92]–[102] XSecuring energy-trading in electric vehicles
individual health-care data and support data-sharing across
XGuaranteeing mutual-confidence among UAVs
different hospitals, medical centers, insurance companies and
patients. During the whole process, the privacy and security of
health-care data can be assured. Furthermore, Sun et al. [90]
put forth an attribute-based signature scheme in decentralized used to deliver product items [98] and acquire real-time traffic
health-care blockchain systems. On one hand, this scheme can flow data [99]. Moreover, the recent study of [100] also shows
verify the authenticity of health-care data and identification of that UAVs can be used to support content-centric networking
the health-care data owner. On the other hand, this scheme can and mobile edge computing. However, it is challenging to
also preserve the privacy of the health-care data owner. The assure the trustworthiness in decentralized non-trusted UAV-
recent work [91] presents an in-home therapy management networks and restrict the misbehaving UAVs [103]. The in-
framework integrating IoT and blockchain-based MEC scheme tegration of blockchain technology with UAV-networks can
to provide secrecy and anonymity assurance. The experimental guarantee the mutual-confidence among UAVs. The work of
results on a prototype demonstrate the effectiveness of the [101] developed an autonomous platform based on Ethereum
proposed system. blockchain to provide the trust-management of UAVs. More-
over, IBM [102] recently applied for a patent to develop
a blockchain-based system to preserve privacy and assure
F. Internet of vehicles and unmanned aerial vehicles security of UAV data. In particular, blocks in blockchains will
Internet of vehicles (IoV) essentially integrates vehicle- store the information related to UAVs including model type,
to-vehicle networks, vehicle-to-roadside networks, vehicle-to- manufacturer, proximity to restricted region. Consequently, the
infrastructure networks and vehicle-to-pedestrian networks. misbehavior of UAVs can be detected and identified in time.
The decentralization, heterogeneity and non-trustworthiness Summary. Table III summarizes major BCoT applications.
of IoV pose the challenges in securing message-transmission In particular, it is shown in Table III that incorporating
and transaction-execution. Integrating blockchain with IoV blockchain with IoT can bring a number of benefits in the
can tackle the above challenges. For example, the work of aforementioned applications. In summary, BCoT has merits
[92] developed a trust-management platform in IoV on top like reducing the cost for trusted third party, assuring security,
of blockchains. In particular, the trustworthiness of messages improving data traceability, verifying the data authenticity and
can be validated via PoW/PoS consensus executed by Road- preserving privacy.
side Units (RSUs). Moreover, blockchain tehcnologies can be
used to protect both the energy and information interactions VII. O PEN RESEARCH ISSUES OF B LOCKCHAIN OF T HINGS
between electric vehicles [93] and hybrid electric vehicles in Although the convergence of blockchain and IoT brings a
smart grids [94], [95]. In the future, incorporating artificial in- number opportunities in upgrading the industry, there are many
telligence, mobile edge computing and blockchain can further challenges to be addressed before the potentials of BCoT can
optimize the resource allocation in IoVs [96]. be fully unleashed. In this section, we identify several major
Recently, unmanned aerial vehicles (UAVs) communication challenges in incorporating blockchain into IoT and discuss
networks can compensate in-sufficient coverage of wireless the potential solutions. Fig. 9 summarizes the open research
communication networks [97]. Meanwhile, UAVs can also be issues for blockchain of things.
14

Fig. 9. Open research issues for blockchain of things

A. Resource constraints [108]. In addition, it is also challenging to manage the keys

Most of IoT devices are resource-constrained. For example, (which are crucial to encryption algorithms) in distributed
sensors, RFID tags and smart meters have inferior computing environment.
capability, limited storage space, low battery power and poor Meanwhile, blockchain systems also have their own security
network connection capability. However, the decentralized vulnerabilities such as program defects of smart contracts [21].
consensus algorithms of blockchains often require extensive In particular, it is shown in [109] that the malicious users
computing power and energy consumption. For example, PoW can exploit Border Gateway Protocol (BGP) routing scheme
in Bitcoin is shown to have high energy consumption [6]. to hijack blockchain messages, thereby resulting in the higher
Therefore, the consensus mechanisms with huge energy con- delay of block broadcasting. The work of [110] also shows that
sumption may not be feasible to low-power IoT devices. a Decentralized Autonomous Organization (DAO) attack stole
On the other hand, the bulky size of blockchain data also $50 million worth of Ethereum by leveraging the vulnerability
results in infeasibility of fully deploying blockchains across of smart contracts.
IoT. For example, the Bitcoin blockchain size almost reaches Potential solutions. Security vulnerabilities of BCoT can be
185 GB by the end of September 2018. It is impossible to fully remedied via either the security enhancement of IoT systems
store the whole blockchain at each IoT device. Meanwhile, the or loophole repairing of blockchain. For example, cooperative
massive IoT data generated in nearly real time manner makes jamming scheme [111] was explored to improve the security
this status quo even worse. Moreover, blockchains are mainly of IoT systems while no extra hardware is required for existing
designed for a scenario with the stable network connection, IoT nodes. Meanwhile, [112] exploits key generations based
which may not be feasible for IoT that often suffers from on reciprocity and randomness of wireless channels in Long
the poor network connection of IoT devices and the unstable Range (LoRa) IoT network. In the perspective of repairing
network due to the failure of nodes (e.g., battery depletion). blockchain loopholes, there are also some advances. In par-
Potential solutions. Incorporating MEC and cloud com- ticular, the recent work of [113] proposes a secure relaying-
puting technologies into BCoT may potentially overcome network for blockchains, namely SABRE, which can prevent
resource constraints of IoT devices. For example, cloud servers blockchain from BGP routing attacks. Regarding DAO attacks,
or some MEC servers may serve as full nodes that store the Corda and Stellar trade the expressiveness for the verifiability
whole blockchain data and participate in most of blockchain of smart contracts [114] so as to avoid DAO attacks.
operations, such as initiating transactions, validating transac-
tions (i.e., mining) while IoT devices may serve as lightweight C. Privacy leakage
nodes that only store partial blockchain data (even hash value Blockchain technologies have some mechanisms to pre-
of blockchain data) and undertake some less-computational- serve a certain data privacy of transaction records saved in
intensive tasks (such as initiating transactions) [104]. The blockchains. For example, transactions are made in Bitcoin via
orchestration of MEC and cloud computing becomes an im- IP addresses instead of users’ real identities thereby ensuring a
portant issue in the sense of allocating resource in BCoT [105]. certain anonymity. Moreover, one-time accounts are generated
in Bitcoin to achieve the anonymity of users. However, these
B. Security vulnerability protection schemes are not robust enough. For example, it
is shown in [22] that user pseudonyms can be cracked via
Although incorporating blockchain technologies into IoT learning and inferring the multiple transactions associated with
can improve the security of IoT via the encryption and digital one common user. In addition, the full storage of transaction
signature brought by blockchains, the security is still a major data on blockchain can also lead to the potential privacy
concern for BCoT due to the vulnerabilities of IoT systems leakage as indicated in [115].
and blockchain systems. Potential solutions. Recently, mixed coins are proposed to
On one hand, there is a growing trend in deploying wireless confuse attackers so that they cannot infer the exact number
networks into industrial environment due to the feasibility of real coins spent by a transaction. However, recent study
and scalability of wireless communication systems. How- [116] demonstrates the weakness of the coin-mixed schemes
ever, the open wireless medium also makes IoT suffering via extensive realistic experiments based on Monero4. More-
from the security breaches such as passive eavesdropping over, the actual transaction can be deduced by leveraging
[106], jamming, replaying attacks [107]. Moreover, due to the vulnerability of the coin-mixed schemes. The work of
the resource constraints of IoT devices, conventional heavy-
weighted encryption algorithms may not be feasible to IoT 4A private digital currency platform (https://getmonero.org/)
 15

[115] presents a memory optimized and flexible blockchain records. However, it often requires the data decryption be-
data storage scheme, which can somewhat reduce the privacy fore conducting data analytics. Nevertheless, the decryp-
leakage risk. tion process is often time-consuming thereby resulting in
the inefficiency of data analytics [124]. It is challenging to
design data analytics schemes on blockchain data without
D. Incentive mechanism in BCoT decryption.
An appropriate incentive mechanism is a benign stimulus Potential solutions. MEC is serving as a crucial complement
to blockchain systems. For example, a number of Bitcoins to cloud computing by offloading computing tasks from distant
(BTC) will be rewarded to a miner who first solves the cloud servers to MEC in approximation to users. As a result,
computationally-difficult task. Meanwhile, a transaction in MEC can improve the response, privacy-preservation and
Ethereum will be charged with a given fee (i.e., gas) to pay context-awareness in contrast to cloud computing. Therefore,
the miners for the execution of contracts. Therefore, there are offloading BDA tasks to MEC servers can potentially solve
two issues in designing incentive mechanisms in blockchains: the privacy-leakage and long latency issue of cloud computing
1) the reward for proving (or mining) a block and 2) the with blockchain [125]. Regarding data analytics on anonymous
compensation for processing a transaction (or a contract). blockchain data, there are some recent advances: 1) complex
However, it is challenging to design a proper incentive network-based community detection [126] to identify mul-
mechanism for BCoT to fulfill the requirements of different tiple addresses associated with an identical user, 2) feature
applications. Take digital currency platforms as an example, extraction of transaction patterns of Bitcoin blockchain data
where miners are keen on the price of digital currency. For to identify payment relationships [127], 3) analysis of user
instance, the BTC reward for a generated block will be accounts and operation codes on Ethereum to detect Ponzi
halved every 210,000 blocks [117]. The reward decrement will fraud behavior [128].
discourage miners to contribute to the solution of the puzzle
consequently migrating to other blockchain platforms. How to F. Scalability of BCoT
design a proper rewarding and publishing mechanism of digital
currency is necessary to ensure the stability of blockchain The scalability of incumbent blockchains also limits the
systems. wide usage of blockchains in large scale IoT. The scalability
of blockchains can be measured by the throughput of trans-
Potential solutions. On the other hand, the reputation and
actions per second against the number of IoT nodes and the
honesty is an impetus to users in private or consortium
number of concurrent workloads [26], [114]. Many blockchain
blockchain systems. Therefore, going beyond digital currency,
systems are suffering from the poor throughput. For example,
reputation credits can be used as incentives in the scenarios
it is shown in [129] that Bitcoin can only process seven
like personal reputation systems [118], sharing economy [119],
transactions per second. In contrast, VISA can process nearly
data provenance [120] and the medication supply chain [121].
2,000 transactions per second and PayPal has the throughput
The recent work [122] presents RepChain, which exploits the
of 170 transactions per second [130], [131]. Ref. [4] shows
reputation of each node to develop the incentive mechanism.
that Bitcoin blockchain may not be suitable for IoT due to
the poor scalability. In summary, the incumbent blockchain
E. Difficulty in BDA in BCoT systems may not be suitable for the applications with a large
volume of transactions especially for IoT.
There is a surge of big volume of IoT data generated in Potential solutions. There are two possible directions in
nearly real time fashion. The IoT data exhibits in massive improving the scalability of blockchains in IoT: 1) designing
volume, heterogeneity and huge business value. Big data more scalable consensus algorithms and 2) constructing private
analytics on IoT data can extract hidden values and make or consortium blockchains for IoT. Regarding 1), we can
intelligent decisions. However, it is challenging for apply choose the consensus-localization strategy to improve the
conventional big data analytics schemes in BCoT due to the throughput of transactions. Meanwhile, we may implement
following reasons: some new blockchain structures such as directed acyclic graph
• Conventional BDA schemes cannot be applied to IoT (DAG) [132] to allow the non-conflicting blocks from the
devices due to the resource limitations. Since IoT devices side-chain to be assembled with the main chain, consequently
have inferior computing capability, the complicated BDA reducing the cost for resolving bifurcation. In addition, we
schemes cannot be deployed at IoT devices directly. may consider integrating PoW with PBFT to improve the
Moreover, the bulky size of blockchain data also leads throughput of PoW similar to Sharding Protocol proposed in
to the infeasibility of the local storage of blockchain data [133], in which less computational-extensive puzzle is first
at IoT devices. Although cloud computing can address solved in PoW and consensus is then reached in multiple small
these issues, uploading the data to remote cloud servers groups.
can also result in the privacy-breach and the long-latency Regarding 2), transactions in private and consortium
[123]. blockchains can be processed much faster than public
• It is difficult to conduct data analytics on anonymous blockchains due to the fully-controlled systems and the limited
blockchain data. Blockchain technologies can protect number of permitted users. Meanwhile, the consensus can
data privacy via encryption and digital signature on data also be easily reached in private and consortium blockchains.
16

Moreover, the fully-controlled blockchains also fulfill the [7] T. M. Fernândez-Caramês and P. Fraga-Lamas, “A review on the use of
requirement that an enterprise needs to have a control on blockchain for the internet of things,” IEEE Access, vol. 6, pp. 32 979–
33 001, 2018.
different strategic sectors, e.g., ERP, MES, PLM and CRM [8] M. S. Ali, M. Vecchio, M. Pincheira, K. Dolui, F. Antonelli, and
systems [59], [114]. Though there are some attempts such M. H. Rehmani, “Applications of blockchains in the internet of
as GemOS [36], Multichain [37] and Hyperledger [38], more things: A comprehensive survey,” IEEE Communications Surveys
Tutorials, pp. 1–42, 2018 (early access). [Online]. Available:
mature private and consortium blockchain platforms serving https://doi.org/10.1109/COMST.2018.2886932
for specific industrial sectors are still expected in the future. [9] A. Panarello, N. Tapas, G. Merlino, F. Longo, and A. Puliafito,
“Blockchain and iot integration: A systematic survey,” Sensors,
vol. 18, no. 8, 2018. [Online]. Available: http://www.mdpi.com/
VIII. C ONCLUSION 1424-8220/18/8/2575
[10] S. Petersen and S. Carlsen, “WirelessHART Versus ISA100.11a: The
The incumbent Internet of Things (IoT) systems are facing a Format War Hits the Factory Floor,” IEEE Industrial Electronics
number of challenges including heterogeneity, poor interoper- Magazine, vol. 5, no. 4, pp. 23–34, Dec 2011.
ability, resource constraints, privacy and security vulnerability. [11] K. Mekki, E. Bajic, F. Chaxel, and F. Meyer, “A comparative study of
LPWAN technologies for large-scale IoT deployment,” ICT Express,
The recent appearance of blockchain technologies essentially 2018.
offers a solution to the issues with the enhanced interoperabil- [12] M. Chen, Y. Miao, Y. Hao, and K. Hwang, “Narrow Band Internet of
ity, privacy, security, traceability and reliability. Things,” IEEE Access, vol. 5, pp. 20 557–20 577, 2017.
[13] O. Khutsoane, B. Isong, and A. M. Abu-Mahfouz, “IoT devices and
In this paper, we investigate integrating blockchain with applications based on LoRa/LoRaWAN,” in IECON 2017 - 43rd Annual
IoT. We name such synthesis of blockchain and IoT as BCoT. Conference of the IEEE Industrial Electronics Society, Oct 2017, pp.
We provide a comprehensive survey on BCoT. In particular, 6107–6112.
we first briefly introduce internet of things and blockchain [14] X. Lu, D. Niyato, H. Jiang, D. I. Kim, Y. Xiao, and Z. Han,
“Ambient Backscatter Assisted Wireless Powered Communications,”
technology. We then discuss the opportunities of BCoT and IEEE Wireless Communications, vol. 25, no. 2, pp. 170–177, April
depict the architecture of BCoT. We next outline the research 2018.
issues in blockchain for next-generation networks. We further [15] J. Zhou, Z. Cao, X. Dong, and A. V. Vasilakos, “Security and Privacy
for Cloud-Based IoT: Challenges,” IEEE Communications Magazine,
discuss the potential applications of BCoT and outline the open vol. 55, no. 1, pp. 26–33, January 2017.
research directions in BCoT. [16] R. Roman, J. Zhou, and J. Lopez, “On the features and challenges of
security and privacy in distributed internet of things,” Comput. Netw.,
vol. 57, no. 10, pp. 2266–2279, July 2013.
ACKNOWLEDGEMENT [17] X. Lu, D. Niyato, H. Jiang, D. I. Kim, Y. Xiao, and Z. Han, “Ambient
backscatter assisted wireless powered communications,” IEEE Wireless
This work was supported by the National Key Re- Communications, vol. 25, no. 2, pp. 170–177, April 2018.
search and Development Program (2016YFB1000101), the [18] J. He, J. Wei, K. Chen, Z. Tang, Y. Zhou, and Y. Zhang, “Multitier fog
National Natural Science Foundation of China (61722214 computing with large-scale iot data analytics for smart cities,” IEEE
Internet of Things Journal, vol. 5, no. 2, pp. 677–686, April 2018.
and U1811462), Macao Science and Technology Development [19] Z. Zheng, S. Xie, H.-N. Dai, X. Chen, and H. Wang, “Blockchain
Fund under Grant No. 0026/2018/A1, and the Program for challenges and opportunities: A survey,” International Journal of Web
Guangdong Introducing Innovative and Entrepreneurial Teams and Grid Services, vol. 14, no. 4, pp. 352 – 375, 2018.
[20] C. Miguel and L. Barbara, “Practical Byzantine fault tolerance,” in
(2016ZT06D211). In addition, this project has also received Proceedings of the Third Symposium on Operating Systems Design
funding from the European Union’s Horizon 2020 research and Implementation, vol. 99, New Orleans, USA, 1999, pp. 173–186.
and innovation programme under the Marie Skłodowska-Curie [21] X. Li, P. Jiang, T. Chen, X. Luo, and Q. Wen, “A survey on the security
grant agreement No 824019. The authors would like to thank of blockchain systems,” Future Generation Computer Systems, 2017.
[22] M. Conti, S. K. E, C. Lal, and S. Ruj, “A Survey on Security and
Gordon K.-T. Hon for his constructive comments. Privacy Issues of Bitcoin,” IEEE Communications Surveys Tutorials,
pp. 1–1, 2018.
[23] B. Chase and E. MacBrough, “Analysis of the XRP Ledger consensus
R EFERENCES protocol,” arXiv preprint arXiv:1802.07242, 2018.
[1] P. Lade, R. Ghosh, and S. Srinivasan, “Manufacturing Analytics and [24] Y. Gilad, R. Hemo, S. Micali, G. Vlachos, and N. Zeldovich, “Algo-
Industrial Internet of Things,” IEEE Intelligent Systems, vol. 32, no. 3, rand: Scaling byzantine agreements for cryptocurrencies,” in Proceed-
pp. 74–79, May 2017. ings of the 26th Symposium on Operating Systems Principles. ACM,
[2] A. Dorri, S. S. Kanhere, R. Jurdak, and P. Gauravaram, “Blockchain 2017, pp. 51–68.
for IoT security and privacy: The case study of a smart home,” in [25] F. R. Yu, J. Liu, Y. He, P. Si, and Y. Zhang, “Virtualization for
2017 IEEE International Conference on Pervasive Computing and Distributed Ledger Technology (vDLT),” IEEE Access, vol. 6, pp.
Communications Workshops (PerCom Workshops), March 2017, pp. 25 019–25 028, 2018.
618–623. [26] T. T. A. Dinh, J. Wang, G. Chen, R. Liu, B. C. Ooi, and
[3] Y. Zhang and J. Wen, “The IoT electric business model: Using K.-L. Tan, “Blockbench: A framework for analyzing private
blockchain technology for the internet of things,” Peer-to-Peer blockchains,” in Proceedings of the 2017 ACM International
Networking and Applications, vol. 10, no. 4, pp. 983–994, Jul 2017. Conference on Management of Data, ser. SIGMOD ’17. New
[Online]. Available: https://doi.org/10.1007/s12083-016-0456-1 York, NY, USA: ACM, 2017, pp. 1085–1100. [Online]. Available:
[4] M. Conoscenti, A. Vetrò, and J. C. De Martin, “Blockchain for the http://doi.acm.org/10.1145/3035918.3064033
internet of things: A systematic literature review,” in 2016 IEEE/ACS [27] G. Zyskind, O. Nathan, and A. Pentland, “Decentralizing Privacy:
13th International Conference of Computer Systems and Applications Using Blockchain to Protect Personal Data,” in 2015 IEEE Security
(AICCSA), Nov 2016, pp. 1–6. and Privacy Workshops, May 2015, pp. 180–184.
[5] M. Banerjee, J. Lee, and K.-K. R. Choo, “A blockchain future for [28] S. S. Chawathe, Clustering Blockchain Data. Cham: Springer Inter-
internet-of-things security: a position paper,” Digital Communications national Publishing, 2019, pp. 43–72.
and Networks, vol. 4, no. 3, pp. 149 – 160, 2018. [29] J. Ream, Y. Chu, and D. Schatsky, “Upgrading blockchains:
[6] A. Reyna, C. Martn, J. Chen, E. Soler, and M. Daz, “On blockchain Smart contract use cases in industry,” Deloitte Press, 2016.
and its integration with IoT. Challenges and opportunities,” Future [Online]. Available: https://www2.deloitte.com/insights/us/en/focus/
Generation Computer Systems, vol. 88, pp. 173 – 190, 2018. signals-for-strategists/using-blockchain-for-smart-contracts.html
 17

[30] N. Szabo, “The idea of smart contracts,” Nick Szabo’s Papers [55] K. Kalkan and S. Zeadally, “Securing internet of things with software
and Concise Tutorials, 1997. [Online]. Available: http://www. defined networking,” IEEE Communications Magazine, vol. 56, no. 9,
fon.hum.uva.nl/rob/Courses/InformationInSpeech/CDROM/Literature/ pp. 186–192, September 2018.
LOTwinterschool2006/szabo.best.vwh.net/smart contracts 2.html [56] P. K. Sharma, S. Singh, Y. Jeong, and J. H. Park, “Distblocknet: A
[31] F. Idelberger, G. Governatori, R. Riveret, and G. Sartor, “Evaluation of distributed blockchains-based secure sdn architecture for iot networks,”
logic-based smart contracts for blockchain systems,” in International IEEE Communications Magazine, vol. 55, no. 9, pp. 78–85, 2017.
Symposium on Rules and Rule Markup Languages for the Semantic [57] I. D. Alvarenga, G. A. F. Rebello, and O. C. M. B. Duarte, “Securing
Web (RuleML). Springer, 2016, pp. 167–183. configuration management and migration of virtual network functions
[32] C. Sillaber and B. Waltl, “Life cycle of smart contracts in blockchain using blockchain,” in NOMS 2018 - 2018 IEEE/IFIP Network Opera-
ecosystems,” Datenschutz und Datensicherheit - DuD, vol. 41, no. 8, tions and Management Symposium, April 2018, pp. 1–9.
pp. 497–500, Aug 2017. [58] I. Afolabi, T. Taleb, K. Samdanis, A. Ksentini, and H. Flinck, “Network
[33] R. Koulu, “Blockchains and online dispute resolution: smart contracts Slicing and Softwarization: A Survey on Principles, Enabling Technolo-
as an alternative to enforcement,” SCRIPTed, vol. 13, p. 40, 2016. gies, and Solutions,” IEEE Communications Surveys Tutorials, vol. 20,
[34] S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” 2008. no. 3, pp. 2429–2453, 2018.
[Online]. Available: https://bitcoin.org/bitcoin.pdf [59] C. Esposito, A. Castiglione, B. Martini, and K. K. R. Choo, “Cloud
[35] “Ethereum: Blockchain APP Platforms.” [Online]. Available: https: manufacturing: Security, privacy, and forensic concerns,” IEEE Cloud
//www.ethereum.org/ Computing, vol. 3, no. 4, pp. 16–22, July 2016.
[36] “GemOS: the blockchain operating system.” [Online]. Available: [60] V. Ortega, F. Bouchmal, and J. F. Monserrat, “Trusted 5g vehicular net-
https://enterprise.gem.co/ works: Blockchains and content-centric networking,” IEEE Vehicular
[37] “MultiChain: Open platform for building blockchains.” [Online]. Technology Magazine, vol. 13, no. 2, pp. 121–127, June 2018.
Available: https://www.multichain.com/ [61] K. Fan, Y. Ren, Y. Wang, H. Li, and Y. Yang, “Blockchain-based
[38] “Hyperledger project,” 2015. [Online]. Available: https://www. efficient privacy preserving and data sharing scheme of content-centric
hyperledger.org/ network in 5g,” IET Communications, vol. 12, no. 5, pp. 527–532,
[39] X. Xu, I. Weber, M. Staples, L. Zhu, J. Bosch, L. Bass, C. Pautasso, and 2018.
P. Rimba, “A taxonomy of blockchain-based systems for architecture [62] C. Chen, M. Lin, and C. Liu, “Edge computing gateway of the indus-
design,” in IEEE International Conference on Software Architecture trial internet of things using multiple collaborative microcontrollers,”
(ICSA), 2017, pp. 243–252. IEEE Network, vol. 32, no. 1, pp. 24–32, 2018.
[40] “Consortium chain development.” [Online]. Available: https://github. [63] N. Abbas, Y. Zhang, A. Taherkordi, and T. Skeie, “Mobile edge
com/ethereum/wiki/wiki/Consortium-Chain-Development computing: A survey,” IEEE Internet of Things Journal, vol. 5, no. 1,
[41] D. Johnson, A. Menezes, and S. Vanstone, “The Elliptic Curve Digital pp. 450–465, 2018.
Signature Algorithm (ECDSA),” International Journal of Information [64] Z. Xiong, Y. Zhang, D. Niyato, P. Wang, and Z. Han, “When Mobile
Security, vol. 1, no. 1, pp. 36–63, 2001. Blockchain Meets Edge Computing,” IEEE Communications Magazine,
[42] K. Christidis and M. Devetsikiotis, “Blockchains and smart contracts vol. 56, no. 8, pp. 33–39, August 2018.
for the internet of things,” IEEE Access, vol. 4, pp. 2292–2303, 2016. [65] M. Liu, F. R. Yu, Y. Teng, V. C. M. Leung, and M. Song, “Computation
[43] Q. Lu and X. Xu, “Adaptable blockchain-based systems: A case study offloading and content caching in wireless blockchain networks with
for product traceability,” IEEE Software, vol. 34, no. 6, pp. 21–27, mobile edge computing,” IEEE Transactions on Vehicular Technology,
2017. vol. 67, no. 11, pp. 11 008–11 021, Nov 2018.
[44] Y. Zhang and J. Wen, “An IoT electric business model based on the [66] P. Yang, N. Zhang, Y. Bi, L. Yu, and X. S. Shen, “Catalyzing cloud-
protocol of bitcoin,” in Proceedings of 18th International Conference fog interoperation in 5g wireless networks: An sdn approach,” IEEE
on Intelligence in Next Generation Networks (ICIN), 2015, pp. 184– Network, vol. 31, no. 5, pp. 14–20, 2017.
191. [67] Y. Dai, D. Xu, S. Maharjan, and Y. Zhang, “Joint load balancing and
[45] M. Massaro, “Next generation of radio spectrum management: offloading in vehicular edge computing and networks,” IEEE Internet
Licensed shared access for 5g,” Telecommunications Policy, of Things Journal, pp. 1–12, 2019 (Early Access). [Online]. Available:
vol. 41, no. 5, pp. 422 – 433, 2017, optimising Spectrum https://doi.org/10.1109/JIOT.2018.2876298
Use. [Online]. Available: http://www.sciencedirect.com/science/article/ [68] Z. Zhou, P. Liu, J. Feng, Y. Zhang, S. Mumtaz, and J. Rodriguez,
pii/S0308596117301416 “Computation resource allocation and task assignment optimization
[46] J. Eggerton, “FCC’s Rosenworcel Talks Up 6G,” https: in vehicular fog computing: A contract-matching approach,” IEEE
//www.multichannel.com/news/fccs-rosenworcel-talks-up-6g, Tech. Transactions on Vehicular Technology, pp. 1–1, 2019 (Early Access).
Rep., September 2018. [Online]. Available: https://doi.org/10.1109/TVT.2019.2894851
[47] R. Saracco, “Let’s start talking about 6G!” http://sites.ieee.org/ [69] Y. Yang, K. Wang, G. Zhang, X. Chen, X. Luo, and M. Zhou,
futuredirections/2018/01/25/lets-start-talking-about-6g/, Tech. Rep., “Meets: Maximal energy efficient task scheduling in homogeneous fog
January 2018. networks,” IEEE Internet of Things Journal, vol. 5, no. 5, pp. 4076–
[48] A. Gatherer, “What Will 6G Be?” https://www.comsoc.org/ 4087, Oct 2018.
publications/ctn/what-will-6g-be, Tech. Rep., June 2018. [70] A. Kusiak, “Smart manufacturing,” International Journal of Production
[49] S. Yrjölä, “Analysis of blockchain use cases in the citizens broadband Research, vol. 56, no. 1-2, pp. 508–517, 2018.
radio service spectrum sharing concept,” in Cognitive Radio Oriented [71] J. Wan, J. Li, M. Imran, D. Li, and F. e-Amin, “A blockchain-based
Wireless Networks. Cham: Springer International Publishing, 2018, solution for enhancing security and privacy in smart factory,” IEEE
pp. 128–139. Transactions on Industrial Informatics, pp. 1–9, 2019 (Early Access).
[50] K. Kotobi and S. G. Bilen, “Secure Blockchains for Dynamic Spec- [Online]. Available: https://doi.org/10.1109/TII.2019.2894573
trum Access: A Decentralized Database in Moving Cognitive Radio [72] I. Konstantinidis, G. Siaminos, C. Timplalexis, P. Zervas, V. Peristeras,
Networks Enhances Security and User Access,” IEEE Vehicular Tech- and S. Decker, “Blockchain for business applications: A systematic
nology Magazine, vol. 13, no. 1, pp. 32–39, March 2018. literature review,” in Business Information Systems, W. Abramowicz
[51] E. H. H. Kure, P. Engelstad, S. Maharjan, S. Gjessing, and and A. Paschke, Eds. Cham: Springer International Publishing, 2018,
Y. Zhang, “Distributed uplink offloading for iot in 5g heterogeneous pp. 384–399.
networks under private information constraints,” IEEE Internet of [73] H. M. Kim and M. Laskowski, “Toward an ontology-driven blockchain
Things Journal, pp. 1–14, 2019 (Early Access). [Online]. Available: design for supply-chain provenance,” Intelligent Systems in Accounting,
https://doi.org/10.1109/JIOT.2018.2886703 Finance and Management, vol. 25, no. 1, pp. 18–27, 2018.
[52] E. Langberg, “Blockchains in Mobile Networks,” https://e.huawei.com/ [74] A. Tapscott and D. Tapscott, “How blockchain is changing finance,”
hk/publications/global/ict$\ $insights/201703141505/, Tech. Rep. 21, Harvard Business Review, vol. 1, 2017.
March 2017. [75] N. Kshetri, “1 blockchains roles in meeting key supply chain manage-
[53] S. He, C. Xing, and L.-J. Zhang, “A business-oriented schema for ment objectives,” International Journal of Information Management,
blockchain network operation,” in Blockchain – ICBC 2018, S. Chen, vol. 39, pp. 80 – 89, 2018.
H. Wang, and L.-J. Zhang, Eds. Cham: Springer International [76] Z. Li, H. Guo, W. M. Wang, Y. Guan, A. Vatankhah Barenji,
Publishing, 2018, pp. 277–284. G. Q. Huang, K. S. McFall, and X. Chen, “A blockchain and
[54] S. Bera, S. Misra, and A. V. Vasilakos, “Software-defined networking automl approach for open and automated customer service,” IEEE
for internet of things: A survey,” IEEE Internet of Things Journal, Transactions on Industrial Informatics, pp. 1–9, 2019 (Early Access).
vol. 4, no. 6, pp. 1994–2008, Dec 2017. [Online]. Available: https://doi.org/10.1109/TII.2019.2900987
18

[77] D. Tse, B. Zhang, Y. Yang, C. Cheng, and H. Mu, “Blockchain applica- [97] Y. Zeng, R. Zhang, and T. J. Lim, “Wireless communications with
tion in food supply information security,” in 2017 IEEE International unmanned aerial vehicles: opportunities and challenges,” IEEE Com-
Conference on Industrial Engineering and Engineering Management munications Magazine, vol. 54, no. 5, pp. 36–42, May 2016.
(IEEM), Dec 2017, pp. 1357–1361. [98] G. Kimchi, D. Buchmueller, S. A. Green, B. C. Beckman, S. Isaacs,
[78] F. Tian, “An agri-food supply chain traceability system for china A. Navot, F. Hensel, A. Bar-Zeev, and S. S. J.-M. Rault, “Unmanned
based on rfid amp;amp; blockchain technology,” in 13th International aerial vehicle delivery system,” 2017, uS Patent 9,573,684.
Conference on Service Systems and Service Management (ICSSSM), [99] L. Wang, F. Chen, and H. Yin, “Detecting and tracking vehicles
2016, pp. 1–6. in traffic by unmanned aerial vehicles,” Automation in Construction,
[79] F. Sander, J. Semeijn, and D. Mahr, “The acceptance of blockchain vol. 72, pp. 294 – 308, 2016. [Online]. Available: http://www.
technology in meat traceability and transparency,” British Food Journal, sciencedirect.com/science/article/pii/S0926580516300887
vol. 0, no. 0, p. null, 2018. [100] N. Cheng, W. Xu, W. Shi, Y. Zhou, N. Lu, H. Zhou, and X. Shen, “Air-
[80] R. Bettı́n-Dı́az, A. E. Rojas, and C. Mejı́a-Moncayo, “Methodological ground integrated mobile edge networks: Architecture, challenges, and
approach to the definition of a blockchain system for the food industry opportunities,” IEEE Communications Magazine, vol. 56, no. 8, pp.
supply chain traceability,” in Computational Science and Its Applica- 26–32, August 2018.
tions – ICCSA 2018. Cham: Springer International Publishing, 2018, [101] A. Kapitonov, S. Lonshakov, A. Krupenkin, and I. Berman,
pp. 19–33. “Blockchain-based protocol of autonomous business activity for multi-
[81] Q. Lin, H. Wang, X. Pei, and J. Wang, “Food safety traceability system agent systems consisting of uavs,” in 2017 Workshop on Research,
based on blockchain and epcis,” IEEE Access, vol. 7, pp. 20 698– Education and Development of Unmanned Aerial Systems (RED-UAS),
20 707, 2019. 2017, pp. 84–89.
[102] A. Kumar, A. Kundu, C. A. Pickover, and K. Weldemariam, “Un-
[82] C. Zhang, J. Wu, Y. Zhou, M. Cheng, and C. Long, “Peer-to-peer
manned aerial vehicle data management,” 2018, uS Patent App.
energy trading in a microgrid,” Applied Energy, vol. 220, pp. 1 –
15/463,147.
12, 2018. [Online]. Available: http://www.sciencedirect.com/science/
[103] B. Li, Z. Fei, and Y. Zhang, “Uav communications for 5g and
article/pii/S0306261918303398
beyond: Recent advances and future trends,” IEEE Internet of
[83] Z. Li, J. Kang, R. Yu, D. Ye, Q. Deng, and Y. Zhang, “Consortium Things Journal, pp. 1–23, 2019 (Early Access). [Online]. Available:
Blockchain for Secure Energy Trading in Industrial Internet of Things,” https://doi.org/10.1109/JIOT.2018.2887086
IEEE Transactions on Industrial Informatics, vol. 14, no. 8, pp. 3690– [104] Y. Dai, D. Xu, S. Maharjan, and Y. Zhang, “Joint Computation Of-
3700, Aug 2018. floading and User Association in Multi-Task Mobile Edge Computing,”
[84] N. Z. Aitzhan and D. Svetinovic, “Security and privacy in decentralized IEEE Transactions on Vehicular Technology, vol. 67, no. 12, pp.
energy trading through multi-signatures, blockchain and anonymous 12 313–12 325, Dec 2018.
messaging streams,” IEEE Transactions on Dependable and Secure [105] T. X. Tran, A. Hajisami, P. Pandey, and D. Pompili, “Collaborative
Computing, vol. 15, no. 5, pp. 840–852, Sept 2018. mobile edge computing in 5g networks: New paradigms, scenarios,
[85] C. Pop, T. Cioara, M. Antal, I. Anghel, I. Salomie, and M. Bertoncini, and challenges,” IEEE Communications Magazine, vol. 55, no. 4, pp.
“Blockchain based decentralized management of demand response 54–61, April 2017.
programs in smart energy grids,” Sensors, vol. 18, no. 1, 2018. [106] X. Li, H. Wang, H.-N. Dai, Y. Wang, and Q. Zhao, “An analytical
[86] K. Wang, Y. Shao, L. Shu, C. Zhu, and Y. Zhang, “Mobile big data study on eavesdropping attacks in wireless nets of things,” Mobile
fault-tolerant processing for ehealth networks,” IEEE Network, vol. 30, Information Systems, vol. 2016, 2016.
no. 1, pp. 36–42, January 2016. [107] J. Lin, W. Yu, N. Zhang, X. Yang, H. Zhang, and W. Zhao, “A survey
[87] C. Esposito, A. D. Santis, G. Tortora, H. Chang, and K. R. Choo, on internet of things: Architecture, enabling technologies, security and
“Blockchain: A panacea for healthcare cloud-based data security and privacy, and applications,” IEEE Internet of Things Journal, vol. 4,
privacy?” IEEE Cloud Computing, vol. 5, no. 1, pp. 31–37, Jan 2018. no. 5, pp. 1125–1142, 2017.
[88] K. N. Griggs, O. Ossipova, C. P. Kohlios, A. N. Baccarini, E. A. [108] Y. Yang, L. Wu, G. Yin, L. Li, and H. Zhao, “A survey on security and
Howson, and T. Hayajneh, “Healthcare blockchain system using smart privacy issues in internet-of-things,” IEEE Internet of Things Journal,
contracts for secure automated remote patient monitoring,” Journal of vol. 4, no. 5, pp. 1250–1258, 2017.
Medical Systems, vol. 42, no. 7, p. 130, Jun 2018. [Online]. Available: [109] M. Apostolaki, A. Zohar, and L. Vanbever, “Hijacking Bitcoin: Routing
https://doi.org/10.1007/s10916-018-0982-x attacks on cryptocurrencies,” in Security and Privacy (SP), IEEE
[89] M. Z. A. Bhuiyan, A. Zaman, T. Wang, G. Wang, H. Tao, and M. M. Symposium on. IEEE, 2017, pp. 375–392.
Hassan, “Blockchain and big data to transform the healthcare,” in [110] S. Adhami, G. Giudici, and S. Martinazzi, “Why do businesses
Proceedings of the International Conference on Data Processing and go crypto? an empirical analysis of initial coin offerings,” Journal
Applications, ser. ICDPA. ACM, 2018, pp. 62–68. of Economics and Business, vol. 100, pp. 64 – 75, 2018.
[90] Y. Sun, R. Zhang, X. Wang, K. Gao, and L. Liu, “A decentralizing [Online]. Available: http://www.sciencedirect.com/science/article/pii/
attribute-based signature for healthcare blockchain,” in 2018 27th S0148619517302308
International Conference on Computer Communication and Networks [111] L. Hu, H. Wen, B. Wu, F. Pan, R. Liao, H. Song, J. Tang, and X. Wang,
(ICCCN), 2018, pp. 1–9. “Cooperative jamming for physical layer security enhancement in
internet of things,” IEEE Internet of Things Journal, vol. 5, no. 1,
[91] M. A. Rahman, M. S. Hossain, G. Loukas, E. Hassanain, S. S. Rahman,
pp. 219–228, Feb 2018.
M. F. Alhamid, and M. Guizani, “Blockchain-based mobile edge
[112] W. Xu, S. Jha, and W. Hu, “Lora-key: Secure key generation system
computing framework for secure therapy applications,” IEEE Access,
for lora-based network,” IEEE Internet of Things Journal, pp. 1–10,
vol. 6, pp. 72 469–72 478, 2018.
2019 (early access).
[92] Z. Yang, K. Yang, L. Lei, K. Zheng, and V. C. M. Leung, “Blockchain- [113] M. Apostolaki, G. Marti, J. Mller, and L. Vanbever, “SABRE: Protect-
based decentralized trust management in vehicular networks,” IEEE ing Bitcoin against Routing Attacks,” in Proceedings of the Network
Internet of Things Journal, pp. 1–10, May 2018 (Early Access). and Distributed System Security Symposium, 2019, pp. 1–15.
[Online]. Available: https://doi.org/10.1109/JIOT.2018.2836144 [114] T. T. A. Dinh, R. Liu, M. Zhang, G. Chen, B. C. Ooi, and J. Wang, “Un-
[93] H. Liu, Y. Zhang, and T. Yang, “Blockchain-enabled security in electric tangling blockchain: A data processing view of blockchain systems,”
vehicles cloud and edge computing,” IEEE Network, vol. 32, no. 3, pp. IEEE Transactions on Knowledge and Data Engineering, vol. 30, no. 7,
78–83, May 2018. pp. 1366–1385, July 2018.
[94] J. Kang, R. Yu, X. Huang, S. Maharjan, Y. Zhang, and E. Hossain, “En- [115] A. Dorri, S. S. Kanhere, and R. Jurdak, “MOF-BC: A memory
abling localized peer-to-peer electricity trading among plug-in hybrid optimized and flexible blockchain for large scale networks,” Future
electric vehicles using consortium blockchains,” IEEE Transactions on Generation Computer Systems, vol. 92, pp. 357 – 373, 2019.
Industrial Informatics, vol. 13, no. 6, pp. 3154–3164, Dec 2017. [Online]. Available: http://www.sciencedirect.com/science/article/pii/
[95] J. Kang, R. Yu, X. Huang, M. Wu, S. Maharjan, S. Xie, and Y. Zhang, S0167739X17329552
“Blockchain for secure and efficient data sharing in vehicular edge [116] M. Möser, K. Soska, E. Heilman, K. Lee, H. Heffan, S. Srivastava,
computing and networks,” IEEE Internet of Things Journal, pp. 1–11, K. Hogan, J. Hennessey, A. Miller, A. Narayanan, et al., “An Empirical
2019 (early access). Analysis of Traceability in the Monero Blockchain,” Proceedings on
[96] Y. Dai, D. Xu, S. Maharjan, G. Qiao, and Y. Zhang, “Artificial Privacy Enhancing Technologies, vol. 2018, no. 3, pp. 143–163, 2018.
Intelligence Empowered Edge Computing and Caching for Internet of [117] K. Saito and M. Iwamura, “How to make a digital currency on a
Vehicles,” IEEE Wireless Communications Magazine, 2019. blockchain stable,” arXiv preprint arXiv:1801.06771, 2018.
 19

[118] A. Yasin and L. Liu, “An online identity and smart contract manage-
ment system,” in Proceedings of 40th Annual Computer Software and
Applications Conference (COMPSAC), vol. 2, 2016, pp. 192–198.
[119] A. Bogner, M. Chanson, and A. Meeuw, “A decentralised sharing app
running a smart contract on the ethereum blockchain,” in Proceedings
of the 6th International Conference on the Internet of Things, 2016,
pp. 177–178.
[120] X. Liang, S. Shetty, D. Tosh, C. Kamhoua, K. Kwiat, and L. Njilla,
“Provchain: A blockchain-based data provenance architecture in cloud
environment with enhanced privacy and availability,” in 2017 17th
IEEE/ACM International Symposium on Cluster, Cloud and Grid
Computing (CCGRID), 2017, pp. 468–477.
[121] D. G. Glover and J. Hermans, “Improving the traceability of
the clinical trial supply chain,” Applied Clinical Trials, vol. 26,
no. 11, pp. 36–38, November 2017. [Online]. Available: https:
//search.proquest.com/docview/1984377517?accountid=28120
[122] C. Huang, Z. Wang, H. Chen, Q. Hu, Q. Zhang, W. Wang, and
X. Guan, “Repchain: A reputation based secure, fast and high incentive
blockchain system via sharding,” 2019.
[123] P. Wang, R. X. Gao, and Z. Fan, “Cloud computing for cloud manu-
facturing: benefits and limitations,” Journal of Manufacturing Science
and Engineering, vol. 137, no. 4, pp. 1–9, 2015.
[124] N. Wang, X. Xiao, Y. Yang, T. D. Hoang, H. Shin, J. Shin, and G. Yu,
“Privtrie: Effective frequent term discovery under local differential pri-
vacy,” in IEEE International Conference on Data Engineering (ICDE),
2018.
[125] Y. Dai, D. Xu, S. Maharjan, Z. Chen, Q. He, and Y. Zhang, “Blockchain
and deep reinforcement learning empowered intelligent 5g beyond,”
IEEE Network Magazine, 2019 (in press).
[126] C. Remy, B. Rym, and L. Matthieu, “Tracking bitcoin users activity
using community detection on a network of weak signals,” in Complex
Networks & Their Applications VI. Cham: Springer International
Publishing, 2018, pp. 166–177.
[127] P. Tasca, A. Hayes, and S. Liu, “The evolution of the bitcoin economy:
Extracting and analyzing the network of payment relationships,” The
Journal of Risk Finance, vol. 19, no. 2, pp. 94–126, 2018.
[128] W. Chen, Z. Zheng, J. Cui, E. Ngai, P. Zheng, and Y. Zhou, “Detecting
ponzi schemes on ethereum: Towards healthier blockchain technology,”
in Proceedings of the 2018 World Wide Web Conference, ser. WWW
’18. Republic and Canton of Geneva, Switzerland: International World
Wide Web Conferences Steering Committee, 2018, pp. 1409–1418.
[Online]. Available: https://doi.org/10.1145/3178876.3186046
[129] K. Croman, C. Decker, I. Eyal, A. E. Gencer, A. Juels, A. Kosba,
A. Miller, P. Saxena, E. Shi, E. G. Sirer, et al., “On scaling de-
centralized blockchains,” in International Conference on Financial
Cryptography and Data Security. Springer, 2016, pp. 106–125.
[130] J. Vermeulen, “Bitcoin and Ethereum vs Visa
and PayPal Transactions per second,” Altcoin Today,
April 2017. [Online]. Available: http://www.altcointoday.com/
bitcoin-ethereum-vs-visa-paypal-transactions-per-second/
[131] S. Albrecht, S. Reichert, J. Schmid, J. Strüker, D. Neumann, and
G. Fridgen, “Dynamics of blockchain implementation-a case study
from the energy sector,” in Proceedings of the 51st Hawaii Interna-
tional Conference on System Sciences, 2018.
[132] Y. Lewenberg, Y. Sompolinsky, and A. Zohar, “Inclusive block chain
protocols,” in International Conference on Financial Cryptography and
Data Security. Springer, 2015, pp. 528–547.
[133] L. Luu, V. Narayanan, C. Zheng, K. Baweja, S. Gilbert, and P. Saxena,
“A secure sharding protocol for open blockchains,” in Proceedings of
the 2016 ACM SIGSAC Conference on Computer and Communications
Security, ser. CCS ’16. New York, NY, USA: ACM, 2016, pp. 17–30.
[Online]. Available: http://doi.acm.org/10.1145/2976749.2978389