Contents Introduction
The number of devices connected to the Internet of Things (IoT) is expected to grow beyond 75 billion by 2025, recording an increase of more than 50 billion devices within the next five years [1]. Internet-connected devices communicating with each other were initially referred to as Machine to Machine (M2M) communication-based device networks [2]. These M2M networks have evolved towards IoT connecting heterogeneous devices that can communicate with each other using various protocols over the internet to exchange a wide range of information [3], [4], [5], [6]. IoT is further enhanced together with the dawn of the Internet of Everything (IoE), which facilitates the interactions among people, devices, and data [7], [8], [9].
The world is experiencing the 4th industrial revolution together with the advancements in IoT and Information and Communication Technologies (ICT) [10], [11]. Moreover, recent developments in 5G mobile communication also facilitates the development of IoT through ultra-Reliable-Low-Latency-Communication (uRLLC) and massive Machine-Type-Communication (mMTC) [6], [12], [13], [14], [15]. Therefore, IoT enabled smart industries are expected to exhibit an exponential growth [16], [17].
The conventional computer-aided industries are getting converted into smart industries due to various reasons such as better quality and productivity, security issues of existing solutions, and ultimately the disruptive nature of smart solutions. So, the industries that become intelligent by adapting smart solutions will evolve while the others will not be able to cope up with the disruptive nature of the smart industries. These smart industries make data driven decision making based on IoT and Big Data Analytics (BDA) concepts [18]. IoT application areas such as agriculture, utilities, healthcare and transportation are expected to be converted into smart IoT applications [19]. Sensor networks have automated and improved the efficiency of agricultural practices [20]. Smart city concept will be supported by the utility meters which will be deployed worldwide as smart IoT systems. Smart healthcare systems will utilize latest technological advancements to perform critical remote surgeries as well. Connected cars is the next revolution in transportation sector.
A. Background
Blockchain is a widely used DLT that is decentralized in nature and continuously grows along with the executed transactions. When a transaction is executed, a new block creation request is initiated. Then, all the nodes in the blockchain network initiate the process of block validation, A validated block will be added to the end of the blockchain by inverse reference pointing to the parent block. This block validation process prevents any alterations to the existing blocks in the blockchain as the malicious party needs to change the relevant block of each node in the network. The key characteristics of blockchain such as decentralization, immutability, trustworthiness, and non-repudiation have made it an ideal candidate for applications that require secure but anonymous and immutable transactions [21]. Due to these characteristics, blockchains have the potential to mitigate the existing issues and revolutionize IoT payments and IoT marketplace concept.
IoT payments - Due to the distributed nature of blockchain technologies, P2P transactions without third-party involvement can be performed. Therefore, the bottleneck of having a central authority due to its cost and performance issues can be mitigated by validating the transactions by a decentralized mechanism. Individual nodes in the blockchain keeps all the committed transaction details and its immutable. The cryptographic mechanisms in blockchain guarantee the integrity and enable secure payments. Also it helps with other issues caused by third parties in IoT financial transactions, namely the lack of anonymity for the users and additional security risks.
IoT marketplace - Global IoT data market value will reach over USD 1 trillion by 2026 [22]. There are several IoT data marketplaces that have already been deployed for selling and buying IoT data. But, the unique characteristics of IoT data have created technical challenges for the success of those platforms. A neutral platform that can be trusted by both the data suppliers and the consumers is essential for the success of IoT market. Blockchain can be used as the base technology to create such a digital trading platform. Further, it can allow IoT device owners to monetize any transaction or exchange data among the devices as a reward instead of a monetary transaction.
The use of third parties for IoT-related financial transactions creates issues such as lack of anonymity and security concerns. When the growing hacking attempts on internet-based platforms are considered, users don’t prefer to share their credit card information or storing their transactions history in IoT marketplaces. When a third party is involved, it leads to an increased transaction fee, also. This is a major concern considering the small-scale micro-transactions that occur in IoT systems. We can use blockchain-based platforms to develop anonymous solutions and remove third parties involved in centralized systems. On the other hand, transaction fees can be an issue in blockchain technology as well. But, the research community is attempting various mechanisms to resolve this issue using the latest DLTs.
The data generated by IoT are tradeable assets that can be even sold to third-party buyers as well. The traditional marketplaces are generally used to share static data. But, the IoT ecosystem requires near-real-time data streams to utilize its actual potential. It requires efficient ways to avoid the initial data consumers from reselling the data without the consent of the data supplier as well. Therefore, mutual trust will be a key component in the IoT marketplace and dynamic trade agreements are required among the parties. The IoT marketplaces do not always consist of trusted parties and smart contracts can be used as a trusted intermediary to create reliable transactions. Datum [28] is a smart contract-based blockchain that provides an option to securely store structured data in decentralized storage. But, this doesn’t address the real-time data requirement of IoT applications. Any IoT data producer should be able to trade the generated data in an IoT marketplace [29].
Blockchain is an ideal candidate to address many of the unique challenges observed in the IoT domain [30]. But, there are obstacles that blockchains in IoT implementation should overcome to be the paradigm shift in the IoT domain. The need to pay a transaction fee to reward miners for their time and efforts is a major obstacle. The market-based transaction fees concept used in typical cryptocurrencies is quite expensive and not suitable for IoT transactions. The transaction time of blockchains is another problematic issue. IoT payment systems require many microtransactions to be completed within seconds. But, public blockchains such as Bitcoin, Ethereum and Ripple require each block be validated by the global blockchain [25]. Bitcoin Lightning Network (BLN), virtualized DLTs (vDLTs) and alternative DLT technologies such as Directed Acyclic Graphs (DAGs) are some examples for research attempts to improve blockchain technology to make it suitable for IoT applications. Even though blockchain needs further optimizations to make it an ideal IoT payment and marketplace platform, the already available research results indicate that blockchain based IoT payment and marketplace concepts can revolutionize the IoT ecosystem.
B. Motivation
As per the summary of surveys given in Table 2, already available surveys discussing IoT payments and IoT marketplaces based on blockchain is very limited. Perera et al. [23] has done a survey of the IoT solutions in the emerging marketplace by discussing and summarising the functionalities provided by each solution.
Enser et. al. [25] examine the compatibility of blockchains for IoT payment transactions by highlighting the characteristics that have made blockchain an ideal solution for P2P transactions between IoT data producers and consumers. Further, this research paper discusses the integration challenges of blockchain to achieve fast and cheap IoT-related transactions as well. Ozyilmaz et al. [31] analyze the positive impact of establishing a decentralized and trustless platform for IoT data sharing. The writers have developed a proof-of-concept data marketplace using Ethereum and Swarm. Shaimaa Bajoudah et al. [32] proposed a decentralized marketplace to trade brokered IoT data. Also, Wiem Badreddine et al. [33] introduce an IoT data marketplace with three different models using MQTT as the publish/subscribe method, Ethereum as the DLT, and Solidity as the smart contract.
C. Our Contribution
Even though, the existing research and survey papers summarize the possible usage of blockchain for IoT, an extensive survey has not been done to research specifically on possible avenues for blockchain to resolve IoT payment challenges and cater marketplace requirements with the expected IoT boost that will be initiated by smart industries.
This paper aims to
Introduce the blockchain-based platforms - The concepts related to blockchain and smart contracts are summarized. Further, the blockchain-based platforms are introduced.
Thoroughly analyze the usage of blockchain for IoT applications - The blockchain-based IoT solutions are discussed.
Examine the IoT payment related issues and the requirement for an IoT market place - The concept of IoT payments and blockchain-based IoT payment solutions are discussed in detail.
Discuss the suitability of blockchain based IoT payment system and IoT marketplace - The concept of IoT market places and blockchain-based IoT market place solutions are discussed in detail.
Discuss technical challenges of IoT - The technical challenges of IoT solutions are introduced and the role of blockchain in solving those challenges are analyzed.
Discuss the integration challenges of blockchain - The unique challenges of blockchain integration are discussed.
D. Outline
The remainder of the paper is organized as follows. Section II is an introduction to IoT and related challenges. Section III presents an overview on blockchain and smart contracts. Section IV summarizes the payment and market place requirements for IoT. How the payment and market place challenges can be addressed by blockchain is discussed in Section V. Available IoT payment systems and IoT market place options based on blockchain are discussed in Section VI and VII respectively. Section VIII introduces the integration challenges of blockchain. Lessons learned and open research issues are discussed in Section IX. Finally, the paper is concluded in Section X.
IoT and Related Challenges
In this section, we discuss the IoT related concepts and the related challenges which have lead to the usage of blockchain-based solutions.
IoT systems connect various smart objects mounted with sensors actuators and software systems. A typical IoT system consists of three main layers including, a perception layer with various IoT devices including sensors and actuators to sense data, a communication layer with various wired and wireless modules to transmit the sensor data, and an industrial application layer with various smart industry verticals. The general IoT architecture which is suitable for describing the latest IoT systems consists of an application layer, cloud computing layer, network layer, fog layer, and object layer is shown in Figure 1. The cloud computing and the fog layers are the major additions to the initial IoT architecture [34].
IoT has the following features [21]:
Heterogeneity of IoT data - IoT systems consist of heterogeneous IoT devices, communication protocols, and IoT data types [35].
Decentralization of IoT systems - IoT systems should be capable to exchange, make use of information and collaborate with each other
Diversity of IoT ecosystems - Sensors, actuators, and software systems used in one IoT system to the other varies a lot.
These characteristics of IoT has resulted in bellow challenges [21]:
Heterogeneity of IoT system – The variation of the devices, communication protocols, etc. have created complex networks and paved the way to many other challenges [35].
Poor interoperability – Due to the decentralized and heterogeneous nature of the IoT systems, it is challenging to exchange information among different IoT systems [36].
Resource limitations of IoT devices – IoT devices consist of limited computing, storage, and power resources. Therefore, the companies are forced to have large-scale cloud infrastructure projects [37].
Privacy and security vulnerabilities – Exchanging IoT data with cloud-based IoT platforms, authentication of vastly decentralized IoT devices using limited resources in IoT devices have created privacy and security concerns [38].
The IoT devices constantly request data from other devices as a service. This ecosystem requires monetary transactions to be performed by sensors and devices in exchange for data services. These transactions and exchanges occur mainly over an internet-based network. Conventional payment models rely on a trusted third-party such as a bank, during transactions. However, considering the number of transactions that can occur with billions of IoT devices and the prevailing trust issues, centralized payment systems will not be able to handle the smart industry predictions. Also, in the IoT ecosystem, devices need to communicate regularly over the internet to exchange information. Malicious attacks from internal or external sources are common during the process. Therefore, at the first gateway to the network, an authentication mechanism that can secure the identity of the IoT devices in the network is essential [39].
The data growth rate in the IoT domain has increased with new devices, sensors, and emerging technologies. The enormous amount of data generated by these IoT applications has enforced companies to deploy large-scale cloud storage solutions. The emerging smart industries need to predict and adapt their solutions based on the insights received from sensor data. Therefore, a trustworthy neutral data-sharing platform is required for the data producers and consumers to trade [31]. These data streams generated by the IoT devices have a resell value to third-party buyers as well. The unavailability of a clearly defined framework for the business side of the IoT ecosystem is negatively impacting its growth. Therefore, a clear business model framework needs to be defined for the IoT data trading.
Blockchain and Smart Contract
In this section, we first discuss the summary of blockchain technology, then summarize the key blockchain characteristics. Further, we introduce the IoT application areas and how blockchain is introduced for those IoT applications.
A. Blockchain Summary
Blockchain is a DLT that keeps a record of transactions in a transparent, auditable, and immutable manner. Block, chain and network are the three core parts of a blockchain. It stores transactions in a chain of blocks while using cryptographic mechanisms [47]. Except for the first block, each block in a blockchain point to its immediately previous block (parent block) using the inverse reference of the parent block (hash value of the parent block) [48]. The first block of a blockchain which is known as the genesis block doesn’t have a parent block. A block structure consists of the following information:
Block size - the size of the block in bytes
Block header - contains block version, a reference to a previous block hash, merkle tree root, mining-related parameters, and nonce
Transaction counter
Transactions
Key characteristics of blockchain include:
Decentralization - The decentralization feature of blockchain distributes the authority among all nodes in the network. This ensures the redundancy of the system when compared with the centralized system approach which requires a trusted third party to operate. High availability of services, improved trust and reduced failure risk are the benifits provided by decentralization.
Immutability - The transactions stored in the ledger are permanent and distributed among the nodes. Therefore, they are unalterable. This immutable nature of blockchain ensures the integrity of the data stored in the ledger [49].
Enhanced security - Due to the decentralized nature of blockchain, any fraudulent party who wants to alter the blocks needs to alter the data stored in all nodes in the network. Encryption mechanisms are used to further enhance security and the cryptographic hash is used in the blockchains as well [50].
Consensus - Conses algorithms are considered as one of the core concepts of blockchain which has made the network being trustless [51]. This is a decision-making process performed by the blockchain nodes in the network to validate the transactions. Even though the nodes in a blockchain don’t trust each other, they trust the transactions validated by the consensus algorithm.
B. Smart Contract
Smart contracts are programs stored in a blockchain that are executed when a predefined condition is satisfied [52]. They execute the terms of a contractual agreement using computerized transaction protocols [53]. The first implementation of a smart contract on the blockchain is Bitcoin. Later, Etherium developed a wide variety of smart contracts on a blockchain [54]. The contractual terms in smart contracts are enforced automatically when given conditions are satisfied. These enforced contractual terms are converted to executable computer programs while preserving the logical statement flows. Once, a smart contract is executed and stored in a blockchain, it is immutable and cannot be modified [21]. Smart contract solutions are used in various IoT applications such as healthcare, manufacturing, and finance [47].
C. Important Blockchain Platforms
A comparison of important blockchain platforms is given in Table 3. Blockchain platforms can be categorized into four different types. They are
Public blockchain - permissionless and anyone with internet access is allowed to become a node in the network which can validate the transactions based on the consensus algorithm used. Eg. Bitcoin and Ethereum
Private blockchain - permission blockchain and network access are restricted only to authorized users. Eg. Multichain and Hyperledger Fabric
Consortium blockchain - this is known as a federated blockchain where multiple organizations can govern the blockchain network. So, this is different from the private blockchain. Eg. R3
Hybrid blockchain - a hybrid of both centralized and decentralized features where some processes are kept public and others private based on the transaction which can be shared in the public network. Eg. Ripple
1) Bitcoin
Bitcoin is a widely used, virtual currency developed based on the concept of cryptocurrency [55]. The validated bitcoin transactions using PoW consensus is updated in a public ledger. Firstly, the transactions generated in a given period will be sorted and stored in a block. Then, the newly generated block will be added to the blockchain using inverse referencing method after it is validated by the other nodes in the public blockchain [56]. Bitcoin operates over a P2P network and it is vulnerable to decentralized network attacks such as the double spending attacks [57].
2) Ethereum
Ethereum is an open-source blockchain solution based on the concept of smart contracts [25]. Ethereum uses a programming language - solidity, decentralized storage service - swarm and the cryptocurrency - ether [31]. It used PoW as the consensus algorithm and later changed to PoS due to the transaction throughput related concerns of PoW [58]
3) Hyperledger Fabric
Bitcoin and Ethereum are famous due to the usage of bitcoin and ether as cryptocurrencies. But, Hypeledger is also gaining popularity, and the software development sector has started using it due to the promising results shown over other competing blockchain platforms [59]. It is an open-source community effort to develop a set of frameworks for industry-level blockchain deployments [60]. Hyperledger fabric is an enterprise-grade DLT platform that is developed to support industry use-cases that require a permissioned network with high scalability and security [61]
4) IOTA
IOTA uses tangle which is a DAG-based DLT technology [62]. IOTA platform is developed to address the IoE requirements by enabling a decentralized IoT data marketplace [31]. IOTA tangle is not a typical blockchain with blocks, chains, and miners. Instead, it uses directed graphs technology to cross-check one another and entangle the steam of transactions together [25]. IOTA platform promises to provide essential characteristics required by IoT.
5) Other Platforms
EOS - EOS is an open-source platform for blockchain-based applications which can be deployed as a public or private network [63]. It uses Delegated Proof-of-Stake (DPoS) consensus algorithm to validate block producers and EOS token as the cryptocurrency [64]. Smart contracts and decentralized autonomous applications (dApps) are also used as core technologies.
Cardano - Cardano is a DLT system that supports smart contracts and dApps. Cardano uses Proof-of-Stake (PoS) algorithm known as Ouroboros [64]. This is the first blockchain platform that was founded based on peer-reviewed research [65].
D. Blockchain for IoT
IoT systems mainly consist of three major components: sensors, computation engine, and actuators. The actual implementation of sensors, computation mechanism, and actuators vary based on the IoT application [66]. The IoT implementations are suffering from technical challenges to achieve their true potential. Blockchain characteristics show a promising future to address the IoT challenges and take IoT applications to the next level. Hence, the convergence of blockchain and IoT has the potential to be the next paradigm in IoT application domain [21]. The Figure 2 summarises the features that will be enabled by blochain for IoT applications.
1) Smart Manufacturing
The conventional manufacturing industry is getting upgraded to smart manufacturing. During a product life cycle, a large amount of data is produced, and BDA-based IoT applications are heavily used [67]. Raw material, Manufacturing equipment, warehouses, and almost everything thing involved in a product life cycle needs to exchange IoT data. Therefore, blockchain-based solutions are suitable for smart manufacturing-related IoT applications and IoT marketplace requirements [68].
2) Smart Agriculture
Smart agriculture applications use technologies such as BDA, cloud computing, IoT sensors to monitor and automate farming operations. Agricultural requirements need IoT applications to install sensors in the land, irrigation systems, weather stations, logistic facilities, and even in seller locations. This is a complex IoT eco-system that can vastly benefit from blockchain IoT applications and IoT marketplace [44], [69], [70].
3) Smart Energy
Smart energy applications are observed in various societal domains such as domestic needs, commercial applications, and other industry requirements. A smart grid is a key smart energy requirement where an electricity network is deployed to detect and react based on energy usage and observed issues to heal them. These applications need to coordinate with multiple parties in the eco-system such as consumers, consumer equipment, power generation, and distribution equipment, administrative and payment systems. Therefore, blockchain-based IoT solutions have the opportunity to resolve the challenges in smart energy applications while creating an IoT marketplace [71], [72], [73], [74], [75].
4) Smart Health
Healthcare-related applications should be dealt cautiously as it involves human life. The healthcare sector includes hospitals, health staff, patients, health equipment, logistics, etc. Therefore, IoT applications need to integrate all these sectors to provide solutions such as patient monitoring, health record maintenance, system automation, and disease predictions. Blockchain-based smart health applications can address strict security concerns related to the healthcare sector and deploy an IoT marketplace solution to allow health data sharing requirements as well [76], [77], [78], [79].
5) IoVs
IoV needs to integrate vehicle networks with other vehicle networks, roadside networks, infrastructure networks, and pedestrian networks to function. Therefore, IoV applications need to exchange sensor data and obtain feedback for users, manufacturers, etc. where an IoT marketplace would also require. Blockchain-based IoV applications have shown promising results to resolve the current IoV challenges and to provide a truly decentralized solution [80].
6) Other IoT Application
The other IoT application domains include transportation, supply chain and retail sector. The concept of IoE has enabled things, processes, data and people to create various IoT domains together. These IoT domains require IoT payments to handle micro-transactions and IoT marketplaces to enable data sharing applications.
IoT applications are expected to be the next paradigm, but the IoT application predictions have not yet been met due to the technical challenges observed during practical implementations. Blockchain-based solutions provide a trustworthy neutral platform for IoT applications to overcome the technical challenges prevailing currently. Blockchain itself needs improvements to handle IoT transaction expectations. But, the current research shows promising results with blockchain-based IoT solutions mainly in the IoT payment and marketplace application domains.
IoT Marketplace
IoT systems use sensors that are capable of sensing the relevant conditions, platforms that can process the collected data, and devices that can function as actuators based on processed data. These IoT eco-systems use cloud computing concepts such as Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure as a Service (IaaS), and emerging concepts such as Sensing as a Service. The IoT platforms require sensor data to function. Therefore, the data consumers are waiting to obtain the sensor data over the internet and they are willing to pay for the obtained sensor data as well.
The concept of the IoT marketplace has become an essential component in modern-day smart application ecosystems. The IoT marketplace value by 2030 is predicted to be over USD 3 trillion [81]. Further, more than one million organizations are expected to use IoT marketplaces to monetize their data. This will generate more than 12 exabytes of IoT data sets for the transactions performed daily [81]. BDEX [82] is a Data as a Service (DaaS) platform established in 2014 to provide actionable data for companies. DAWEX [83] is another global data exchange platform established in 2017 to monetize the generated data.
An IoT data marketplace mainly consists of three components. They are data suppliers, data consumers and the platform to handle the data marketplace.
Data suppliers - They are the data owners. IoT devices such as smart home appliances, smart watches, weather stations and health monitoring devices are a few examples for the commonly used IoT data suppliers which generate data.
Data consumers - They are the parties who are willing to purchase and use the data generated by the data suppliers. Research institutions,smart applications such as IoV and marketing agencies are a few examples for IoT data consumers.
Data platform - It is required to handle the enormous amount of data generated by IoT sensor applications and share the data among the data consumers. Both the data consumers and suppliers need to register in this platform. Then based on the requirement of the data consumer, the platform should handle any relevant transaction fee and distribute data as per pre-defined policies that can be agreed by both the parties.
Figure 3 shows an IoT data marketplace architecture that can be used to share the generate IoT data among the required consumers. The registered IoT sensors generate IoT data and act as the data suppliers. The type of IoT data generated and the relevant compensation details must be updated during the registration process. The consumers need to compensate the suppliers and obtain the generated data.
A. Key Challenges in Current IoT Marketplaces
Data consumers and data suppliers seamlessly trade in an IoT marketplace. This allows companies to make use of both publicly available data and privately-owned data for their use cases. But IoT when creating an IoT marketplace, below mentioned challenges can be observed.
Lack of cooperation among IoT platforms - It is challenging to exchange data between different IoT systems, due to the inherent nature of being decentralized and heterogeneous. Hence, the interoperability of IoT platforms is challenging to be achieved while results in data silos.
Requirement of a trusted third party - Currently, the normal practice is to compensate the IoT data suppliers using a trusted third party such as a bank. When the micro-transactions that occur in the IoT ecosystem are considered, transaction fees and the centralized mechanism are not efficient approaches.
IoT devices specific technical challenges - Limited resource availability is a common feature in IoT systems. Hence, the typical authentication and authorization approaches cannot be implemented and the systems are vulnerable to security threats.
B. Role of Blockchain for IoT Marketplaces
The contributing parties have multi-dimensional benefits from a decentralized IoT data marketplace.
1) Technical Benefits
No need to maintain backend IoT platforms - The IoT data marketplace will act as the backend to integrate the sensors and the IoT applications.
Availability of a vast pool of IoT data - IoT applications have insights from a pool of sensors. So, individual data silos with limited insights aren’t getting created.
Optimized software for IoT sensors and applications - The IoT data marketplace is responsible for integrating the consumers and the suppliers. Therefore, the relevant software code or other application configurations that should run in sensors and IoT applications can be standardized.
Actionable insight reselling capability - The data consumers can obtain raw data and further improve them into actionable insights and sell them to other interested third parties. Since we have a common IoT data marketplace, it is possible to compensate the original data suppliers.
2) Economical Benefits
IoT data monetization for data suppliers - New set of business models involving sensor data generators, data consumers will create more diversified IoT application use cases.
Reselling IoT data with added value - The data obtained by data consumers can be resold after converting them into actionable data insights.
IoT data economy - Even though it is predicted that IoT applications will revolutionize the world, only data silos are getting created. An IoT data marketplace will allow the true potential of IoT data monetization to be achieved.
The functional block diagram of a blockchain-based IoT marketplace varies based on the technologies used and a general block diagram is given in Figure 4. Let’s analyze the already deployed similar solutions based on previous research.
Kazım Rıfat Özyılmaz et al [31] introduced a decentralized and trustless data marketplace platform for non-real-time and non-critical IoT applications to store and access IoT data. They highlight the fact that on-chain or off-chain data storage mechanism, monetization method and suitable tools and capabilities to create an IoT platform are the key considerations, when creating an IoT marketplace. The blockchain used is Ethereum and the decentralized storage used is Swarm. Smart contract is used for transparent data collection sharing to implement the decentralized IoT data marketplace. It is important for the data consumers to know the geolocations. Hence, GeoHex is used to facilitate easy querying of geolocations. A validation and feedback mechanism is used to rank the quality of IoT data, based on consumer feedback. Ether is used as the monetizing method. A off-chain scaling solution called payment channels (Raiden for Ethereum) is used to support instant transactions.
Gowri Sankar Ramachandran et al [84] propose a decentralized marketplace for smart cities. Sellers with data products and buyers interested in their data are fundamental components in the data marketplace. They have used smart contracts to register sellers and post data products for the buyers to search and find them. In this implementation, Ethereum blockchain is used along with an off-chain distributed file storage system named InterPlanetary File System (IPFS). Meta-data organization is proposed to follow any standard JSON format. Streaming Data Payment Protocol (SDPP) is used to enable real-time micropayments. The rating process of buyers and sellers in the marketplace is implemented using a smart contract as well.
Pooja Gupta et al [85] proposed a three-tiered architecture consists of participants of the marketplace, facilitators to supervise service areas, and regulators to ensure that the facilitators are adhering to the privacy regulations. IoT devices are resource-constrained, so facilitators ease the burden on those devices by acting as fog nodes. They used BigChainDB, a decentralized database system to maintain IoT system data. A blockchain named Martchain is built using Ethereum and smart contracts is executed to automate the trading. A watermarking technique is used to identify reselling of data and compensate the data suppliers. The regulators form a consortium blockchain named policy-chain to validate the compliance of facilitators by executing a policy contract, a smart contract.
Wiem Badreddine et al [33] developed a blockchain-based IoT data monetization framework. They use Message Queuing Telemetry Transport (MQTT) protocol and Ethereum smart contract to build three different real-time IoT data sharing solutions. The overall system architecture consists of a smart contract, publishers who provide IoT data, subscribers who consume IoT data, and an untrusted MQTT broker to connect the publishers and subscribers. They propose three solutions; Trace-MAX provides maximum traceability by asking the participants to write detailed information in the distributed ledger. Trace-MIN is the minimum traceability solution that required only the brokers to record minimal information allowing only a basic trace. Trace-BF uses bloom filters to manage data hashes with fewer verification operations on the blockchain.
IoT Payments
In an IoT ecosystem, based on the application the devices use various technologies to generate, communicate and analyze data. The interaction among people. devices and data are called the Internet of Everything (IoE) [25] nowadays. IoE needs the ecosystem to perform monetary transactions in exchange for services over the internet. The introduction of 5G technology is expected to further grow the potential use-cases of IoE.
IoT payment requirements are generated by various IoT verticals. Some of the example scenarios would be a smart car authorizing fuel payments as it approaches the fuel station, a smart health device sharing the prescription with a smart pharmacy, and proceeding with the drug purchase, and a smart agricultural applications ordering and paying for fertilizer and water supplies. M2M transaction value is expected to reach USD 27.62 billion by 2023 [86].
Smart cities - The concept of cashless cities can connect any object to smart city network and eliminate the inefficiencies of payment mechanisms that can be observed in current systems. A research done by Visa in 100 cities shows that smart payment mechanisms can provide USD 470 billion in direct net benefits per year [87].
Smart homes - Modern day houses contain devices with embedded financial functionalities. Automated utility payments and automated supply ordering are a few examples where IoT has enabled pay-per-use business models and service offerings.
Smart transport - The latest connected cars have inbuilt payment functionalities for everything from gas to parking and infotainment. Similar functionalities are introduced to the whole ecosystem of smart transportation. As an example, Honda and Visa together have introduced infotainment system that allows users to pay gas, parking, food, etc [87].
Smart retail - IoT payments have created a subdivision of commercial activities where unattended retail is becoming more common. Amazon’s new cashless, cashier-less stores which allow customers to collect items off shelves and automatically get charged upon exiting. According to new estimates, this has generated more than 50 % more revenue on average than typical convenience stores [88].
Smart grids - The power production, transmission and distribution is monitored, controlled and smart concepts such as smart metering and peer to peer energy trading are introduced with smart grid concept [89].
Smart agriculture - This domain mainly deals with buyer-seller relationship in many areas including purchasing seeds, sharing climate data, selling crops, agriculture field monitoring, etc. Therefore, IoT related micro-payments which can vastly benefit by introducing a distributed payment mechanism based on blockchain is an integral part of the eco-system [69]
Smart healthcare -Smart healthcare systems deals with sensitive patient information and deals with various payments requirements with patients, hospitals, healthcare staff, pharmaceutical industries etc. [90]
Supply Chains - Supply chains related transactions are subjected to issues such as money laundering, sanction violations, bribery, etc. Therefore, blockchain-based IoT payment solutions can improve the efficiency and reduce frauds [91].
A. Key Challenges in Current IoT Payments Systems
Conventional IoT applications deploy their own IoT ecosystem with sensors and actuators with network connectivity and IoT platform to run the relevant control algorithms to run the application. This has created IoT network silos for individual applications. As a result, the capital cost of IoT application deployments is high and the widespread use of IoT solutions is hindered.
Requirement of a trusted 3rd party - Traditional IoT business models rely on a trusted third party to perform financial transactions. An intermediary such as a bank is used during the process and the actual advantages of P2P transactions are not allowed to be obtained because of this.
Lack of anonymity for the users - IoT application users are reluctant to share their payment information and allow the transaction history to be maintained in IoT applications.
Requirement of micro-payments - IoT deals with micro-transactions and in order to perform a single IoT transaction, multiple sub-transactions might need to be performed. The conventional payment mechanisms are not designed to handle such micro-payments and using a trusted third party has increased transaction fees as well.
Issue of fiat currency - Even in today’s world, the paper-based currency is used for 85% of all global transactions. This is common in developed economies as well. The usage of fiat currency in the United Kingdom, United States, and Germany are recorded as 48%, 55%, and 67% respectively [92]. The usage of physical fiat currencies is a major blocking point for the advancement of the IoT sector. The cost of producing, printing, and securing fiat currency, the inefficiency of payments, unsuitability to digital economic needs are some of the major concerns of physical fiat currencies.
Static agreements on payment rates - IoT requires to perform dynamic micro-transactions [93]. Due to traditional business models that are prevailing in the industry, IoT platforms have to enter into static agreements with IoT data suppliers. But, the data requirements of IoT platforms are dynamic in nature. So, the latest IoT trends urge for solutions that has the capability to handle dynamic payment terms.
B. Role of Blockchain for IoT Payments
The key characteristics of blockchain include decentralization, immutability, and enhanced security. These characteristics make the blockchain technology ideal for IoT payments. Since the blockchains are distributed, direct P2P transactions can be performed without the need of a trusted third party. The IoT payments can be made unalterable due to the immutability provided by the consensus mechanism and smart contracts used by blockchains. Therefore, the IoT transaction history can not be modified or reversed.
Worldwide people have adopted electronic payments. The consumer identity theft attempts are common and current payment system ecosystems such as payment cards are vulnerable to such instances. The process of tokenization converts the sensitive data into non-sensitive tokens. Therefore, sensitive data can be secured as the original data gets replaced with an unrelated value. Blockchain-based distributed payment systems can protect the users from identity thefts as payment tokens can easily be used in the blockchain payment ecosystem [94].
The conventional fiat currency faces issues in currency issuance, payment methods, and currency storage requirements. Therefore, digital currency usage is ivolving and the acceptance of bitcoin by El-Salvador in June 2021 as legal tender, is an example for the possibility of world moving towards cryptocurrencies. Xuan Han et al [95] have analyzed some of the major cryptocurrencies and proposed an scheme to use digital currency during their research work.
The eco-system of blochkchain-based IoT payments is given in Figure 5. The blockchain-based decentralized IoT payment schemes will provide various benefits to IoT payment domain.
No dependency with third-party payment platforms - The traditional business models rely on a trusted third party to proceed with payment transactions. The most widely used mechanism is to use a third-party such as a bank to proceed with transactions. Blockchain-based IoT payment options have enabled actual P2P transactions without any dependency on a third party.
Anonymous transactions - When a third party such as a bank is involved in the transaction process, the users have to provide sensitive information such as credit card information to proceed with payments. If IoT data from multiple platforms is required, the users have to expose their sensitive payment information multiple times. Considering the security risks involved in the current digital era, this is not preferred by IoT data consumers. So, using blockchain, a certain amount of anonymity can be maintained.
Less transaction fee - In traditional E-business models, the transaction fee is a major issue as micro-transactions are a unique characteristic of IoT. Blockchain-based solutions are improved and necessary research is being done to minimize the transaction fee. Also, IOTA-tangle-based payment approaches have introduced payment options without any transaction fee.
Technical Challenges of IoT Payment and Marketplaces
The traditional IoT business models used for IoT payments and marketplaces rely on central authority. This has created technical challenges such as IoT data security and privacy issues, creation of data silos without any collaboration, requirement of a central payment systems, lack of anonymity and fradulent transaction attempts. Table 4 shows how the existing IoT payment and marketplace challenges are addressed by blockchain solutions and Table 5 highlights how IoT challenges are addressed by blockchain in real world implementations.
A. IoT Data Security
1) Technical Challenge of IoT Data Security
Most of the IoT devices are controlled remotely over the internet to achieve the desired functionality. These devices use standard communication protocols to share information among the devices in the IoT ecosystem via communication networks. The IoE concept highlights the smart connected things concept where each device is having some sensor module to collect data and a communication module to connect to the IoT application network. These IoT devices include home appliances, health monitoring devices, weather stations, tracking devices, and much other equipment based on the IoT application domain. The data acquired from the end devices need to be shared in real-time with the IoT application platforms. Considering the critical nature of the IoT ecosystem, enabling IoT data security is a critical factor for the development of all IoT fields.
The IoT deployment architectures need to be secure from privacy, integrity, and confidentiality-related security attacks. IoT ecosystem consists of inter-connected networks which inherit the security issues of computer networks. The heterogeneous devices are having power, memory and other resource constraints that make it further challenging to address the security issues through complex mechanisms [26].
2) Role of Blockchain in Solving IoT Data Security Issue
Blockchain solutions are introduced for tracking and monitoring products, goods, and assets to increase trust and security. They maintain the integrity of the distributed transactions. Therefore, IoT devices can be registered with a defined set of attributes in a distributed ledger such as blockchain and resolve the existing security and trust issues of IoT platforms.
The Blockchain IoT (BIoT) concept in [39] explains how sensor data-related transactions can be included in a blockchain. This provides essential security features such as publishing sensor data in distributed ledgers, the immutability of the records, authentication, and non-repudiation of data. They are identified as solutions with protection from data tampering and usage of compromised IoT devices, secure communication, user authentication, and trustworthiness [96].
Even though blockchain-based solutions can establish IoT security, they are also vulnerable to security issues. If the randomness of private keys is limited, then they can easily be compromised by the attackers. Further, transaction privacy and other security threats such as double spending attacks need to be carefully analyzed when introducing a blockchain solution [27].
3) Summary
Even though blockchain systems offer a robust approach for IoT security, they are also vulnerable. Based on the security threat, the research community has suggested different frameworks. But, a single framework that is resilient against many combined attacks and feasible to implement is a research challenge. The fate of blockchain-based security in the era of quantum computing is yet to be properly analyzed.
B. IoT Data Privacy
1) Technical Challenge of IoT Data Privacy
The IoT data produced within the IoT ecosystem need to be transmitted, processed, and stored securely without compromising privacy. It is common for IoT data to contain sensitive information including personal data. User data should not be disclosed without obtaining consent from the data owners. IoT ecosystems are complex, decentralized, and consist of heterogeneous system elements. Therefore, it is challenging to preserve the privacy of IoT data.
The privacy concerns related to IoT data negatively affect the adoption of it as users are reluctant to use a system that is not capable of respecting the privacy requirements of the users. The traditional authorization protocols such as Role Based Access Management (RBAC), OAuth 2.0, and OpenID are too complex to run in most of the resource constraints IoT environments. User privacy is considered a major concern by the regulators as well. Therefore, it is essential to guarantee the privacy of IoT data [26].
2) Role of Blockchain in Solving IoT Data Privacy Issue
Smart contracts can be executed in blockchains to provide IoT device authentication in the decentralized system and they are less complex compared to the traditional authentication protocols. Blockchain solutions have their own privacy-related concerns. One of the main issues that should be addressed to preserve data privacy is not letting and sensitive data be lost or erased from the system [97]. Mixing technique and anonymous solutions are the two existing solutions for privacy protection [98]. Both centralized mixing and decentralized mixing models have their own disadvantages. In centralized mixing, the transaction time is a major concern, and in decentralized mixing increased complexity is a major concern [98]. The use of cryptography-based techniques to preserve privacy and transaction time is a common approach to preserve the privacy.
Smart contracts can be used to define access rules, access time periods, or other required conditions to ensure data privacy and the user right levels can also be managed [26]. Tiffany Hyun-Jin Kim et al [99] developed a mechanism named Self-Sovereign Privacy (SSP) to protect the privacy and integrity of the data collected by IoT devices. This method removed the risk of having a single point of failure and minimized the cryptographic operations that must be performed on IoT devices. Fitwi et al. [100] Fitwi et al. [100] proposed a blockchain-based privacy protection scheme for surveillance cameras to perform surveillance activities by capturing videos without compromising user privacy.
3) Summary
Privacy of IoT data needs to be preserved to enhance user trust in the IoT ecosystem. Blockchain solutions provide possible solutions to address IoT data-related privacy concerns. Even though blockchain is an ideal candidate to address many of the IoT-related technical challenges, widespread adoption of blockchain-based technologies is still hindered due to the privacy concerns of blockchain itself. In public blockchains, user information shared is disclosed to all nodes. Further, the other types of blockchains are also facing data privacy-related concerns where research community is trying to provide an ideal solution.
C. Lack of Cooperation Among IoT Platforms
1) Technical Challenge of Not Having Cooperation Amoing IoT Platforms
It is challenging to exchange data within the IoT ecosystem due to the decentralized nature and heterogeneity of IoT systems and it is challenging to achieve interoperability as well. The distributed IoT resources make it difficult to manage and the distribution of smart objects of multiple IoT applications at the same location makes it more complex. Therefore, the cooporation among IoT platforms is important to reduce complexity and deployment costs [101].
Let’s consider an example: An agricultural IoT system needs weather data for analytical purposes. The weather stations established by the meteorology department independently collect weather data for weather predictions. Since the two IoT platforms are working independently without any cooperation, the agricultural IoT system will need to deploy a separate weather station. If there is a common IoT marketplace, agricultural IoT platform and weather station platform can exchange information and achieve resource optimization. The different IoT verticals need to cooperate with each other to prevent the issue of small silo networks without much use is getting created.
2) Role of Blockchain in Providing Cooperation Among IoT Platforms
Blockchains can keep the transactions in an immutable manner and the transaction records in the blockchain are transparent and reliable. Further, data monetization can be enabled via cryptocurrency-based transactions without the need for fiat currency. Therefore, the blockchain-based IoT data marketplace concept can deploy a platform for various IoT applications to coordinate with each other.
Let’s consider an example: A smart health monitoring device fixed on an athlete can monitor the health conditions. Once the device owner advertises the data via an IoT marketplace, hospitals, AI-based training institutions, pharmaceutical companies, and any other third-party system can obtain the stream of data to be used in their IoT applications. This type of approach can allow the IoT platforms to coordinate with IoT platforms and fast track the IoT adoption.
Dai et al. [21] highlighted the convergence of blockchain and IoT and proposed an interoperable IoT platform architecture. Abou-Nassar et al. [109] introduced a decentralized and interoperable trust model based on blockchain for healthcare-based IoT.
3) Summary
Due to the lack of cooperation among IoT platforms, small-scale silo networks are getting created. This is a major concern for the widespread adoption of IoT solutions. Blockchain provides a promising solution to the IoT platform interoperability issue via the blockchain-based IoT platforms. But, the integration challenges of blockchain such as performance and scalability issues observed with IoT platforms that generate a high volume of data need to be resolved with further research.
D. Requirement of Central Payment Systems
1) Technical Challenge of Having a Central Payment System
The conventional IoT business models rely on a trusted third party to act as an intermediary to perform monetary transactions. This is a major limiting factor to prevent the adoption of true P2P applications. The recent boom in IoT and the expected growth in the IoT ecosystem suggest that IoT applications and relevant monetary transactions will increase exponentially. In an IoT data marketplace, it is preferred to allow data suppliers and consumers to perform the transactions in a decentralized manner [33].
2) Role of Blockchain in Implementing Decentralized Payment Solution
Due to the distributed nature of blockchains, they allow direct P2P transactions without the need of a trusted third party. Blockchain technologies bring capabilities such as data tracking, coordinating, and allowing a large number of devices to be handled without a centralized approach. The parties in a blockchain might not trust each other. But, the immutability of blockchains has enabled the parties to trust each other without a central authority. From the past, there were proposals for decentralized P2P Wireless Sensor Networks (WSN) [24]
P2P transactions without third-party intervention can be allowed in a distributed payment system [110]. Chanthong et al. [111] designed and built an electronic payment system for electric vehicle (EV) charging using blockchain and smart contract technologies to control and manage payments and to decentralize the payment system. Dimitriou et al. [112] developed a data payment and transfer scheme that uses bitcoin payments to reward users for detailed electricity measurements they submit to a utility provider (UP) and other applications such as crowdsensing. Even though the technology is promising which has hindered the usage of blockchain as a decentralized payment platform. The technology has its own scalability issues. Payment channel networks have been proposed to increase transaction throughput and decrease transaction confirmation latency, [113].
3) Summary
IoT deals with micro-transactions. Each transaction of cryptocurrency requires a certain amount of computation and attracts transaction fees. Therefore, the computation and transmission overheads are a major concern in blockchain-based decentralized payment systems.
E. Lack of Anonymity
1) Technical Challenge of Lack of Anonymity
Users do not want the IoT applications to store their payment and transaction history details. Eg: Imagine a farmer who’s using third-party weather station data to control an irrigation system. This farmer has to pay for the weather station data, smart irrigation system, etc. So, the farmer’s bank information data need to be shared with the third parties who are handling these IoT platforms. If the use case involves more IoT platforms, the number of times the bank information to be shared will also increase.
2) Role of Blockchain in Providing Anonymity
Blockchains excel in anonymity when compared to traditional centralized IoT platforms. Gordon et al. [114] used pseudonymization which consists of removing some of the information necessary to identify an entity. The research conducted by Kshetri et al. [115] proposed a blockchain-enabled e-voting (BEV) system to allow the eligible voters to anonymously cast their vote using a computer or smartphone. Lin et al. [116] introduced Decentralized Conditional Anonymous Payment (DCAP). It is difficult to regulate Decentralized Anonymous Payment (DAP) systems. Therefore, the anonymity feature of blockchain can be exploited by criminals for money laundering and other cybercrime.
3) Summary
Blockchain has the capability to provide anonymous solutions via decentralized payment systems. But, complete anonymity can be criminally exploited. Therefore, the decentralized payment systems should be designed after carefully considering the level of anonymity to be allowed based on regulatory requirements to avoid criminal exploitation and reasonable privacy protection.
F. Other Technical Challenges
1) Technical Challenge of High Transaction Fee
If a third-party organization is used in IoT solutions, a separate transaction fee should be paid for their service. Since IoT payments include many micropayments, the higher transaction fee is a major concern for IoT payments and market places [93]. Therefore, a payment that is both small and metered is required for IoT systems. We can use cryptocurrencies to remove the fiat currency usage requirement, Further, the decentralized solutions, remove the need for a third party to perform monetary transactions. Therefore, the merchant payment fees can be reduced and users can receive funds immediately via cryptocurrencies. The transaction fee is an issue in blockchain technologies as well, but there is multiple research that has been done to improve this significantly.
2) Static Agreements Issue
Due to the current static agreements, there is a possibility for the data owner to loose the ownership of data due to reselling possibility. An example scenario would be a small house generating sensor data and providing them to marketplace, then the data can be resold by-passing the owner.
IoT Applications
Blockchain technology can be applied in many IoT domain applications. Blockchain technologies can be applied in different IoT domains as shown in Figure 6. These IoT applications include autonomous vehicles, smart agriculture, smart cities, smart grid, and smart trading. A summary of related research work in IoT application domains are mentioned in Table 6.
A. Autonomous Vehicles
1) Introduction
IoV is considered as an emerging concept in Intelligent Transportation Systems (ITS) which has integrated the Vehicular Adhoc Network (VANETs) to IoT [117]. The overall IoV ecosystem that consists of smart vehicles is interdependent with communication networks, and external environments including roads, traffic lights, road signs, pedestrians, and all relevant domains relevant to transportation. IoV is a major component in the concept of smart city as well [118].
2) Role of Blockchain-Based Solution
IoV has integrated smart vehicles with the internet. The overall IoV ecosystem includes smart vehicles with sensors, road network, pedestrians, other vehicles, and other infrastructure. Therefore, in order to guarantee road safety, a common information exchange platform is required among the IoV ecosystem. ITS needs this platform to be secure, trustworthy, and immutable to achieve the intended objectives [117]. IoV faces technical challenges such as effectively using the scare spectrum, allocating channels for communication, and utilizing transportation infrastructure appropriately based on the traffic conditions [119]. IoV applications can resolve these challenges by integrating blockchain with cryptographic techniques and edge computing. Innopolis University in Russia implemented a M2M billing service for electric autonomous vehicles. Their primary focus was to introduce a solution based on IOTA’s tangle network for the M2M monetary transactions that need to be performed by the vehicles with charging stations for electricity consumption. The introduced payment framework acts as a meter to exchange IOTAs for the consumed power based on the number of kWh [25].
3) Summary
IoV aims to establish a novel and secure smart vehicle ecosystem. IoV is still trying to resolve many security and privacy vulnerabilities. The blockchain-based solutions that are emerging in IoV are capable of resolving the technical challenges of IoV by enabling secure data communication. Secure IoV communications can utilize methodologies such as the High-Performance Blockchain Consensus (HPBC) algorithm. But, the number of transactions required to update blockchain ledgers poses serious issues for vehicles as these may consume the available energy.
B. Smart Agriculture
1) Introduction
Smart agriculture is a revolutionary concept that has allowed farmers to access real-time crop data very easily and respond accordingly. Farmers can analyze data and make informed decisions rather than relying on their gut feeling. Efficiency in all aspects of farming is critical to get the maximum yield and meet the increasing demand. Smart agriculture expects to improve all food supply chain elements by eliminating the middleman and deploying a transparent and efficient system. The application requirements of agriculture supply chain include query efficiency, security and privacy, the authenticity and reliability of data [120].
2) Role of Blockchain-Based Solution
In the context of building an inter-organizational mechanism of data sharing and value creation, blockchain technology is believed to be the favorable candidate, compared to many other information and communication technologies (ICT). The need for a middleman can be eliminated by introducing transparent and efficient blockchain-based food supply chain solutions. This allows the buyers to track the origin of the product, product delivery time, and even the environmental conditions of the field as well [69]. Mohsin Ur Rahman et al. [44] proposed a distributed data sharing system for smart agriculture which consists of four main components namely smart agriculture, smart contract, Interplanetary File System (IPFS), and agriculture stakeholders. Lu et al. [70] designed a blockchain-based agricultural data sharing model and system architecture.
3) Summary
Smart agriculture applications need systems that can guarantee query efficiency, security, and privacy, authenticity, and reliability of data. Blockchain-based smart agriculture platforms are capable of deploying a transparent and trusted ecosystem where farmers can have access to instant agriculture-related data such as the seed quality, climate environment-related data, payments, soil conditions, and crop market status.
C. Smart Cities
1) Introduction
Smart cities are developed as a solution to the growing urbanization challenge to achieve sustainable development goals. They are a combination of smaller smart networks. These data-driven services are developed based on ICTs and the data is acquired via mechanisms such as sensors, cameras, human inputs, crowd-sourced data from mobile phones and vehicles. The data sources might be wholly owned by local authorities, a single organization, or any heterogeneous group of individuals and organizations. The data owners must be compensated by the data consumers in a smart city for sharing the data with them [84].
2) Role of Blockchain-Based Solution
A smart city consists of various parties that need to sell, find and buy data to function their applications. The data suppliers need to be compensated for sharing information. Therefore, the IoT data marketplace platforms are essential in smart cities. Blockchain’s unique characteristics make it an ideal candidate for such an IoT marketplace.
El Majdoubi et al. [122] developed a smart blockchain-based solution to preserve privacy and security in a smart city environment. Stefano Loss et al [123] demonstrated how a blockchain-based platform can be used to handle land registration. Xie et al. [124] reviewed how blockchain technology is applied in various domains of smart cities including smart citizens, smart healthcare, smart grid, smart transportation, and supply chain management. Hakak et al. [125] did a case study on a conceptual blockchain-based architecture that can secure a smart city.
3) Summary
A smart city is an ecosystem of various IoT applications such as transportation, healthcare, energy, manufacturing, education, administration, and logistics. The characteristics of blockchain such as transparency, automation, decentralization, and immutability are helpful in achieving the ultimate smart city concept.
D. Smart Grid
1) Introduction
Smart grids are the core of smart energy solutions. It allows the power generation sources and the power consumers to exchange information and deliver energy in an automated and distributed network. Micro-grids are deployed within small communities using renewable energy and Energy Storage Systems (ESS). They act as a platform to trade locally generated energy with each other in the community [71].
2) Role of Blockchain-Based Solution
Dekhane et al. [72] presented a blockchain-based power distribution ecosystem for smart cities. This system used a wallet-based currency, called “Green Coin”, for power transactions such as buying, selling, and lending. Hamouda et al. [73] presented the development and case-study validation of a comprehensive transactive energy market framework with linked blockchain and power system. Hussain et al. [74] developed an ethereum-based blockchain solution for energy trading. The research results are shown for a case where energy transactions are undertaken between the Distribution System Operator (DSO) and the smart meters of individual houses. Ha et al. [75] designed a power smart contract system based on blockchain for renewal energy trading market that consists of power producers and consumers (Prosumers).
3) Summary
The concept of smart energy aims to integrate green and renewable energy technologies into the conventional power grids efficiently. The blockchain-based typical power trading markets consist of power producers and power consumers who are capable of producing renewable energy. Blockchain and smart contract technologies used in energy trading helps to implement a distributed network with transparency and immutability
E. Smart Healthcare
1) Introduction
Rapid growth is observed in the usage of wearable bio-sensors and smart healthcare-related use cases. The enormous amount of data produced by the healthcare ecosystem need to be shared among the parties involved. This ecosystem will eventually improve the health condition of people by using their own data as it provides an overall better understanding of a patient rather than relying only on Electronic Health Records (EHR). In the current context, due to the hospital or specific patient-specific IoT solutions, only health data silos are getting created. Therefore, it is required to enable data sharing among both private and public healthcare sector applications to prevent the less useful data silos from getting created. [126].
2) Role of Blockchain-Based Solution
The main requirement of a health marketplace is to create a data-centric decentalized ecosystem and use AI applications to further improve the solutions. Decentralized marketplaces are implemented based on secure smart contracts and DLTs. This has enabled data producers to transact with data consumers while maintaining anonymity. Blockchain-based platforms can create a distributed and trusted user network for data sharing [76]. BCT has the potential to enhance data-sharing due to its properties such as transparency, traceability, and immutability [77].
Alsharif et al. [78] proposed a decentralized blockchain-based medical data marketplace for the medical record sellers to sell their data to interested buyers. Kumar et al. [79] established a hyperledger fabric-based blockchain network among patients and medical institutions to share patients’ data securely and reliably. Jaiman et al. [76] developed a blockchain-based data-sharing consent model using smart contracts to access control the health data. Dubovitskaya et al. [77] presented a systematic literature review to analyze the motivations, advantages, limitations, and future challenges faced when applying the distributed ledger technology in oncology.
3) Summary
Smart healthcare applications consisting of various elements such as health monitoring devices, research institutes, patients, and various other sections that need to share health-related information with each other. Considering the sensitive nature of health data, health data sharing should be allowed based on individual consent. But, resource limitation is a major bottleneck. Further, BCT uses crytographic techniques to guarantee data privacy and security.
F. Content Trading
1) Introduction
BitTorrent technology is an option for content data sharing. But, it infringes on the copyright, which taints its public perception [127]. The art market performs USD 200 billion worth annual trading, It is one of the largest unregulated markets in the world which is responsible for one-third of crimes committed [128].
2) Role of Blockchain-Based Solution
The blockchain-based digital content marketplace can resolve the issue of digital content trading issues that are observed in current market environments. The decentralized market which will be created using blockchain allows the users to engage in content publishing, hosting, accessing, downloading, and paying activities.
Heo et al. [129] proposed a new blockchain system named the Secret Block-based BlockChain (SBBC). This uses both off-chain and on-chain network components, to solve the issues observed with the blockchain system. Khan et al. [130] proposed a blockchain solution based on Ethereum to protect content and transactions. A blockchain-based art trading system named ArtChain was introduced as a pilot project in [121].
3) Summary
Current content trading applications need solutions for the existing problems such as copyright violations, forgery, and falsification in the digital content trading environments. Even though blockchain is an ideal solution for the issues in the content trading domain, due to the scalability concerns, it is difficult to propagate digital content to the blockchain network. Therefore, blockchain-based content trading platforms consisting of off-chain and on-chain network components are introduced as a solution.
Integration Challenges of Blockchain
The term blockchain is a leading buzzword in modern day tech world. But, the actual implementations of blockchain technologies are hindered due to various technical limitations mentioned in Table 7 which the research community has been trying to resolve for the last few years. The issues of transaction fee an transaction time are major challenges. The scalability issues of blockchain with respect to transactions and storage are also having a negative impact towards the blockchain adaption. Lack of standards, new security and privacy issues, latest quantum resistance related concerns, intermittent connection issues and other challenges such as regulatory resistance are challenges that should overcome for the widespread use of blockchain technologies in the real world applications such as IoT marketplaces.
A. Issue of Transaction Fee
1) Introduction
Micro-payments are an essential component in IoT applications as they constantly request data as a service from IoT devices. These transactions need to be compensated with some sort of small, metered, and anonymous payment option. The transaction fee is a major bottleneck in blockchain-based IoT application solutions. Some blockchain solutions require the transaction validators to be compensated. The transaction fees are market-based, which makes them quite expensive.
2) Possible Solutions
The on-chain solution is a method to increase scalability by modifying only elements within a blockchain. The transmission cost is lower than the conventional way in this method. But, increasing the block size is not a scalable solution. The off-chain solutions improve scalability by processing the transactions outside the blockchain and solve the transaction cost issues as they are handled outside the blockchain [131]. DAG-based approaches, aiming to provide cheap blockchain services with low latency and high throughput, are emerging as a solution to blockchain transaction fee issue [132].
Fehnker et al. [133] analyzes the Bitcoin Unlimited (BTU) in which the transmission limit is higher and the transmission cost is lower than the conventional way. Guo et al. [134] provided a detailed analysis of the Bitcoin Lightning Network (LN). The LNs use an off-chain mechanism to reduce transaction fees. Pierro et al. [135] analyzed the influence factors on Etherium transaction fee. Yang et al. [136] proposed a DAG-based blockchain for resource-constrained VSNs.
3) Summary
In blockchain, once a transaction is created, the user must pay the transaction fee to the minor. This transaction fee is a major issue as light transactions such as micro-payments are a regular occurrence in IoT. On-chain and off-chain solutions are proposed as solutions to current blockchain scalability and transaction fee issues. But, handling the micro-transactions generated by various IoT systems is still an issue. Research community is using DAG-based solutions for various IoT applications as a latest solution.
B. Scalability Issue: Transactions
1) Introduction
The PoW concept introduced in initial blockchain applications has the core concept of competing for computation power [137].In PoW, the members in the blockchain need to solve complex problems, purely as a need for evidence, but not as a real requirement for a solution. The resource requirement of this process cannot be catered by IoT devices [138]. So, the transaction time is a problematic issue for many blockchain-based technologies.
2) Possible Solutions
Unlike PoW, PoS uses coin age i.e. the contribution to the blockchain network which doesn’t require a high computational power [139]. Hafid et al. [140] did a comprehensive survey regarding the existing solutions to blockchain scalability. Hazari et al. [141] proposed a method using parallel mining which accelerated the process of PoW when compared with solo mining as the maximum of two miners will try to solve a specific block at a given time. Also, in literature, the solutions related to blockchain scalability can be classified into first layer solutions such as sharding, bigger blocks and DAGs and second-layer solutions such as payment channels and side chains [140].
Both PoW and PoS consensus mechanisms avoid forking and maintain a single version of blockchain ledger by slowing down the access rate of new blocks [138]. The use of DAGs is proposed as a solution to decrease transaction time and improve scalability [132]. Instead of arranging the nodes in linear chains, they are arranged as a directed graph in DAGs. The consensus in DAG-based solutions before creating a new node is usually reached by confirming a given number of previous transactions. Hence, it eliminates the requirement of complex consensus mechanisms, ultimately improving the scalability. Yang et al. [142] introduced CoDAG which improved the linear structure of traditional blockchain protocol using DAG. CoDAG achieved a throughput of 394 TPS, which is higher than Bitcoin and Ethereum.
3) Summary
The blockchain scalability issue in terms of transactions is a major issue that the research community has been trying to resolve so it will be suitable for IoT applications. We can classify those solutions into two categories as first layer solutions which modify the structure of the blockchain and second layer solutions which include more transactions in a block to increase throughput. The main chain’s structure is not changed in second-layer solutions. Therefore, network security is not sacrificed when compared with first-layer solutions. The research community is interested in combining both first and second-layer solutions to achieve higher throughput which is essential for adopting blockchain-based solutions for IoT applications.
C. Scalability: Storage
1) Introduction
Storage capacity and scalability have been deeply questioned in the blockchain. The scalability of storage capacity is another major bottleneck in the blockchain. In typical blockchains, the chain continues to grow. Especially the full nodes need significant storage to store the complete chain. Nodes require more resources as they grow and this reduces the scalability of the system and it can impact the system performance. For example, when the Ethereum blockchain platform is considered, it requires all the nodes in a network to participate in the validation and if the data volume in the application is significant, this results in considerable processing delay [143].
2) Possible Solutions
Distributed storage systems can store a large amount of data off the chain. Therefore, it is proposed to combine blockchain with existing distributed storage systems [124]. Block compression techniques are used to reduce some redundant data of a block that has been already stored. Blockchain pruning is used to remove non-critical historical data from the blockchain while preserving the security [144].
The Storage Compression Consensus (SCC) algorithm used by Kim et al. [145] compresses a blockchain in each device to prevent storage capacity limitations in lightweight IoT devices. Kim et al. [146] proposed a selective compression scheme using a checkpoint-chain to prevent the limitation of accumulating the compression results which needs to validate the retained blocks. Sohan et al. [147] proposed a distributed storage system IPFS is used to bypass the storing liabilities and to increase throughput. Matzutt et al. [148] developed a scheme named Coin-Prune to prune old-blocks.
3) Summary
Each node in a traditional blockchain needs to process and store the complete transactions to the genesis block. The IoT devices have limited computing and storage resources. Therefore, blockchains cannot be used with most IoT applications due to this limitation. Combining blockchains with existing storage systems allows the solutions to store a large amount of data off the chain.
D. Lack of Standards
1) Introduction
The term “blockchain” has been one of the most widely used tech buzzwords and it is evolving rapidly when compared with the attempts to introduce a standardization framework. So, the emerging blockchain technologies are used without proper standardization provided by a recognized international organization [149]. This is causing regulatory bodies not to accept blockchain-based technologies as they might not interoperate with each other and it will be difficult to integrate with traditional information systems as well [150].
2) Possible Solutions
A major standardization initiative was initiated on blockchain and distributed ledger technologies through a Technical Committee of the International Organization for Standardization, ISO/TC 307. Other organizations like IEEE1 and the International Telecommunication Union (ITU) have also established standardization efforts on blockchains to identify the needs and responsibilities of their members and stakeholders as users, developers, and operators of this new technology [150].
International Organization for Standardization (ISO) is currently working on 15 work programmes related to the standardization of blockchain technologies and distributed ledger technologies [151]. International Telecommunication Union (ITU) established Application of Distributed Ledger Technology (FG DLT) in May 2017 to identify and analyze DLT-based applications and services related to telecommunication use cases and concluded their work in August 2019 [152]. ITU-T work programme SG20 which is responsible for studies relating to the Internet of things (IoT) and its applications are currently working on blockchain-based applications and frameworks [153].König et al. [149] referred to local and international standardization organization’s publications and provided a set of comparison criteria for future work and a comparison of the existing standards work.
3) Summary
Standardization improves interoperability and provides a clear view of technical aspects for the industry. The existing issue of lack of standardization and clarity is a major obstacle to the adoption of the technology. Therefore, the standardization of blockchain technology is one of the major steps towards enabling interoperability and obtaining regulatory acceptance of the technology. But, introducing standards to an emerging technology can limit its advancement as well.
E. New Security and Privacy Issues
1) Introduction
Blockchain-based IoT systems use encryption and authentication strategies to ensure the security of data. Even though these strategies protect the transaction security of blockchains, the privacy of blockchain-based IoT systems has always been a major concern [154]. Some of the blockchain vulnerabilities are listed below.
Liveness attack: The confirmation time of a target transaction is delayed in a liveness attack. The three phases of liveness attack are the preparation phase, transaction denial phase, and blockchain retarder phase [59].
51% vulnerability: 51% attack can be performed and the entire blockchain can be controlled when a single miner’s hashing power is greater than 50% of the total hashing power of the entire blockchain [155], [156].
Double Spending (DS) attack: When the proportion of computing power possessed by an attacker is higher than that of the honest network, DS attacks can be performed [157]
Selfish mining: In selfish mining, malicious nodes don’t immediately disclose their newly mined blocks and deflect their behavior from the standard pattern [158].
Smart contract vulnerabilities: If an attack based on a smart contract is successfully executed, it will cause the smart contract to perform in an expected manner and result in losses to the parties involved [159].
2) Possible Solutions
The Conflux is a high throughput and fast confirmation blockchain platform which uses a novel consensus protocol to secure against double-spending attacks and liveness attacks [160]. Jang et al. [157] analyzed profitable DS attacks and guide how to set a block confirmation number for a safe transaction.
Singh et al. [27] did a comprehensive survey on the security attacks, challenges, and solutions for the distributed IoT networks. Li et al. [59] did a survey on the security of blockchain systems. Chicarino et al. [158] presented a simple heuristic to detect the presence of selfish mining attacks in PoW based blockchain networks. Sayeed et al. [159] proposed an attack categorization for smart contract vulnerabilities and highlighted the flaws in existing vulnerability detection methodologies. Li et al. [160] developed, high throughput and fast confirmation blockchain platform named the Conflux.
3) Summary
Blockchain solutions also face major security concerns. For example, transaction malleability is one of the main security issues and it is caused by delayed information in the hash transaction, DoS attacks are also a common phenomenon, there are attacks that affect privacy and confidentiality of data. A security attack on a critical IoT application can result in major losses. Therefore, when adopting blockchains for IoT applications, its security issues should be addressed properly after analyzing the similar solutions discussed in this research.
F. Other Challenges
1) Intermittent Connection
In today’s context, many of the services rely on continuous network connectivity. Therefore, communication infrastructure has become a crucial factor. However, 100% population coverage of networks is not available in all areas and even in the areas with coverage, intermittent connectivity is a common issue. Blockchain solutions require continuous network connectivity to constantly exchange data with its peer nodes [161].
A blockchain-based payment scheme can be built using smart contracts, a token-based admission control, account management, and mining rewards distribution for intermittently connected regions [161]. Yining et al. [161] proposed a blockchain-based digital payment scheme that can deliver reliable services on top of unreliable networks in remote regions using Etherium. Xiao et al. [155] proposed an analytical model which can assess the impact of network connectivity on the PoW blockchain and its impact on consensus security.
Lessons Learned and Future Research
In this section, We outline the lessons learned and future research challenges, along with available requirements to improve blockchain-based IoT payment and marketplace solutions.
A. Technical Challenges
1) Lessons Learned
The traditional e-business models used in IoT related payment and marketplaces have hindered the development of IoT applications. IoT data security and privacy are major concerns when transmiting and storing IoT data. A central payment system managed by a third party is required for IoT solutions and lack of anonymity, high transaction fee, fradulent transaction attempts and static agreements are some of the major technical challenges observed in such solutions.
2) Open Research Problems
How to overcome IoT data security and privacy issue?
What’s the best solution to decentralize the IoT platform solutions?
How to improve anonymity in IoT solutions ?
How to resolve the transaction fee and scalability hurdles in IoT solutions?
3) Preliminary Solutions
Data security and privacy can be provided to IoT solutions by using blockchain-based solutions with smart contract and cryptopraphic techniques. Blockchain is a DLT by nature, so central authority can be removed. So, the lack of anonymity and transaction fee related issues can also be overcome by blockchain-based solution which remove the third party from IoT solutions.
4) Future Research Directions
Research community have proposed various DLT and DAG based solutions to address the technical challenges that have limited the true potential of IoT. A single blockchain-based solution for all IoT technical challenges or introducing standardized blockchain-based solutions to resolve specific IoT technical challenges as a guideline will help the development of IoT domain immensely by overcoming its technical challenges. The right balance between blockchain and other cryptographic techniques to ensure data security and privacy need to be identified.
B. Applications
1) Lessons Learned
IoT systems connect various smart objects mounted with sensors actuators and software systems which can sense and collect information from the physical environment and then take necessary actions on them. The IoT ecosystem needs a way for sensors and devices to make monetary transactions in exchange for services. Therefore, IoT payment and marketplace requirements exist in various IoT application domain such as autonomous vehicles, smart agriculture, smart cities, smart grid, smart healthcare and content trading. Due to the characteristics such as decentralization, immutability, non-repudiation, traceability and trust of the BCT, they can be applied in all these IoT application domains.
2) Open Research Problems
How can the blockchain technology along with modern cryptographic techniques, and edge computing be used in IoT applications ?
What’s the best blockchain-based platform for IoT marketplace requirement of various IoT applications?
How can IoT application data silos be prevented ?
3) Preliminary Solutions
Blockchain-based solutions such as HPBC are integrated into secure IoV communications. Various decentralized data marketplaces, focusing on the product smart contract and querying components are introduced in smart cities. Prosumer energy sharing applications use decentralized BCT solutions. Smart health applications use BCT-based data sharing approaches. Content trading is an emerging IoT application where BCT-based solutions were introduced.
4) Future Directives
The IoT domain mainly deals with micro-transactions and transaction fee and scalability issues of blockchain are major bottlenecks for IoT applications to use the BCT. Therefore, when developing an IoT marketplace solution for any IoT application domain, latest DLT-based plaforms such as hyperledger fabric or DAG-based platforms such as IOTA need to be considered. Since, IoT applications deal with sensitive information, ensuring privacy in blokchain-based solutions with right balance of BCT and typical cryptographic techniques is a must.
C. Integration Challenges
1) Lessons Learned
Even though blockchain is highly regarded by research community for IoT applications, the actual implementations of blockchain technologies are hindered due to various technical limitations. The transaction fee that has to be paid to miners is a major obstacle in blockchain-based IoT applications as micro-transactions are a common occurence in IoT domain. The scalability in terms of transactions and storage is another major concern for adopting blcokchain-based solutions. Emerging blockchain technologies are used without proper standardization and this has resulted in lack of trust regarding overall blockchain solutions. Privacy and latest security threats such as liveness attack, 51% vulnerability DS attack and smart contract vulnerabilities need to be addressed. The quantum resistance has threatened core blockchain security implementations which are based on cryptographic and has function related mechanisms. Intermittent connectivity issues observed in rural areas and regulatory concerns can also be considered as significant blockchain integration challenges.
2) Open Research Problems
How to reduce the transaction fee required to reward the miners?
How can the scalability issue of blockchain be resolved ?
How to standardize the emerging blockchain technologies to gurantee interoperability ?
How can blockchain survive the privacy and new security threat ?
Can blockchain survive quantum resistance ?
Will the rejection of blockchain based solutions by regulatory bodies hinder its development effort ?
3) Preliminary Solutions
The proposed on-chain solutions tries to resolve the scalability issue by modifying only elements within a blockchain while the off-chain solutions improve the scalability by processing the transactions at outside the blockchain. The emerging DAG-based approaches are suggested for IoT applications as a solution to blockchain transaction fee issue. ISO and ITU has initiated standardization related directives. Novel conses mechanisms are tested to address blockchain security and privacy issues. The post quantum schemes are adapted by blockchain-based solutions to face quantum resistance.
4) Future Directives
The blockchain solutions are vulnerable to various security threats and a standard framework need to be introduced for the application developers to follow, so the vulnerabilities can be minimized. The research community need to analyze the quantum resistance capability of their blockchain-based solutions and improve the solutions to face the inevitable future. The lack of standardization is a main reason for regulatory resistance and this is a major drawback for the widespread usage of blokchain-based IoT application solutions. Hence, a standardization approach similar to the mobile communication technology standards need to be followed for blockchain as well.
D. Emerging Directions
1) Quantum Resistance
The cryptographic techniques such as Rivest, Shamir, Adleman (RSA), and Elliptic Curve (EC) are used to secure blockchains by protecting stored data [162]. The latest developments in the quantum computing domain can cause security issues that have never been considered before. For example, Shor’s algorithm running on a powerful quantum computer can break the public-key algorithms in polynomial-time [163]. These public-key algorithms can be broken in polynomial-time with Shor’s algorithm on a sufficiently powerful quantum computer [163]. Therefore, the usage of both public-key crypto systems and hash functions is threatened by the evolution of quantum computing.
pre-quantum and post-quantum cryptosystems are introduced to withstand the quantum attacks. For example, Code-based - to support error correction codes, multivariant-based - to solve the complex multivariate equations, lattice-based - rely on n-dimensional spaces with a periodic structure, Isogency-based - to withstand the quantum attacks on elliptic curves [163].
Gao et al. [164] defined post-quantum blockchain (PQB) and proposed a secure cryptocurrency scheme by combining the lattice-based signature scheme with blockchain together. Fernández-Caramès et al. [163] reviewed some of the post-quantum schemes and analyzed their application to the blockchain. A development framework for a scalable, quantum-secured permissioned blockchain named Logicontract (LC) was introduced by Sun et al. [165].
The use of blockchain and other DLTs for numerous applications has evolved significantly in the last few years. The main reason for the widespread research interest in blockchain is its characteristics that we discussed in section 2 due to public-key cryptography and hash functions. The fast progress of quantum computing which uses methodologies such as Grover’s and Shor’s algorithms will be an immense threat to core concepts of blockchain such as public-key cryptography and hash functions. Therefore, the research community needs to closely follow the developments in quantum computing and redesign blockchains to withstand quantum attacks.
2) AI ML Integration
IoT has been a most widely used technology in the recent past for various application domains. The IoT characteristics and its issues such as security concerns and micro-transactions have paved the way to utilize blockchain, AI and ML to introduce technological improvements in the IoT application domains [166]. Due to the rapid adoption of IoT, an exsessive amount of IoT data is getting generated. Therefore, big data and AI, along with blockchain is vital to accurately analyze IoT data in real-time [167]
Unal et al. [168] introduces how blockchain with Federated Learning (FL) can create a secure big data analytics service for IoT. Shahbazi et al. [169] applied an integrated methods of blockchain and ML to create a smart manufacturing system. Supriya et al. [170] reviewed how the utilization of ML, big data, and blockchain technology is important for the health sector advancements.
Conclusion
The adoption of IoT applications has hindered due to technical limitations inherited by IoT. These include data security and privacy issues, lack of cooperation among IoT platforms, the requirement of a central payment system, and lack of anonymity. Therefore, IoT applications require a platform with characteristics such as decentralization, immutability, enhanced security, and anonymity to handle their micro-transactions efficiently and cost-effectively. This paper analyzes the technical challenges faced by IoT applications and the suitability of blockchain-based solutions to address those challenges. Especially, the IoT payment transactions and data sharing marketplaces can be deployed by using blockchain-based technologies. In this research paper, we have identified the IoT application areas such as smart health, smart agriculture, IoVs, smart manufacturing, and content trading as some of the major IoT application domains that have shown promising results with blockchain-based solutions. Even though blockchain-based solutions and DAG-based solutions show promising results, they also have integration challenges such as the issue of the transaction fee, scalability issues in terms of transactions and storage, security and privacy issues, and the latest challenge: quantum resistance that needs to be addressed before being the next revolution of IoT to achieve its true potential.