Research Report | International Exchange Blockchain Application Survey


Since the advent of Bitcoin, there have been many alternative cryptocurrencies and blockchain solutions on the market. The blockchain technology has undergone major evolutionary steps, and also paved the way for more global interconnection solutions in the coming years. This article will start from the original design of the blockchain-international exchange, and try to comprehensively introduce and analyze the case and application of the blockchain in international exchange.


Introduction to Blockchain Development

First generation blockchain technology

Bitcoin is the first digital currency to solve the double payment problem (Note 1). It does not require trusted authorities or centralized servers. Later, the concept of blockchain was separated from its implementation in Bitcoin. Blockchain has the function of acting as a distributed ledger, tracking and recording ownership exchange, which makes it a broader application scenario for this basic technology in addition to digital currencies. Bitcoin's design inspired other applications and played a vital role as a relatively large proof of concept. Due to its scalability limitations and slow processing speed, from the perspective of the payment network, Bitcoin has a large shortage, but as a groundbreaking application, this is a revolutionary breakthrough.

Note 1: Double payment is a concept of a digital currency failure mode, that is, the same digital currency can be spent more than twice.

Second generation blockchain technology

In just a few years, the second generation of blockchain technology came into being. The second generation blockchain is a network on which developers can build applications. The most typical application is the Ethereum platform. Ethereum is a distributed computing platform based on open source public blockchain with smart contract functions. Ethereum provides a foundation layer in the form of a built-in Turing complete programming language, which facilitates the creation of any application. Any user can write smart contracts and decentralized applications with custom specifications, formats, and rules.

Ethereum provides a decentralized virtual machine that can execute computer programs using a global network of nodes. Vitalik Buterin introduced Ethereum in a 2013 white paper with the goal of building distributed applications. The platform went live only two years later. Ethereum attracts a large and dedicated supporter, developer, and corporate community. The contribution of Ethereum as a second-generation blockchain is to expand the technology's capabilities from the original database that supports Bitcoin to more general-purpose platforms that run decentralized applications and smart contracts.

As of 2019, Ethereum is the largest and most popular platform for building distributed applications. Everything from social networks to financial applications exists on the Ethereum platform. The development of blockchain technology is moving towards the development of a global distributed cloud computing platform on which any application can run at the scale and speed of today's major websites.

Third generation blockchain technology

The current technology still faces major scalability issues related to throughput and processing speed. Solving these issues is the core of the third-generation blockchain technology and a necessary condition for large-scale applications such as international exchange. The high energy consumption during mining (Note 2) cannot be adopted on a large scale. Ethereum requires each node to publicly store all smart contracts on the blockchain. A growing data link will slowly degrade the performance and speed of the platform. In response to these problems, many technologies have entered a stage with potential solutions.

Note 2: Mining is a nickname for exploration methods to obtain digital currencies such as Bitcoin. It is named because it works very similarly to mining minerals. In addition, Bitcoin prospectors who perform mining work are also called miners.

Lightning Network

The Lightning Network is an off-chain method to solve the scalability problem of Bitcoin. For small payments, Bitcoin transaction fees may be greater than the transaction amount. Lightning Network solves this problem by establishing a payment channel between the two parties. When setting up a shared channel, the Lightning Network will use a multi-signature address and a Bitcoin buffer. The channel contains a balance sheet, which is updated and signed every time they want to make a transaction without involving the blockchain. Both entities can close the channel at any time by broadcasting only the last signed balance sheet to the Bitcoin network, which will release funds accordingly. They can limit each transaction to two, instead of conducting a bitcoin transaction for each small purchase, once to open the channel and once to close the channel. Only the latest signed balance sheet can be used to unlock funds. The system does not require a direct payment channel between the parties. If these channels are already established, payments can be made between intermediate products. Therefore, the Lightning Network attempts to minimize the complexity of the network to minimize the number of channels and intermediary equipment required to conduct these transactions. In the 20 months since Lightning Network launched its autonomous network, it now has a total of 32,000 channels, 4800 nodes with active channels, and a total capacity of US $ 8.6 million.

Altcoins-RIPPLE XRP and Stellar Lumen XLM

The alternative coin launched after Bitcoin's success is called Altcoins. Stellar Lumen XLM and Ripple XRP are two trending cryptocurrencies, and their goal is to become the future of the banking industry. Both of them offer fast cross-border transactions at extremely low fees. These services will eventually be integrated into more banking systems and will serve as a currency interface between banks, companies and people. Ripple focuses on inter-bank relationships and has gained the trust of many companies and banks. Ripple's founder left the company in 2014 and then continued to develop Stellar, a branch of Ripple. Stellar is a completely decentralized, non-profit branch of Ripple with a vetted consensus algorithm. Stellar's main goal is to pay in small amounts, and it also provides a free-to-use service. Unlike Ripple, Stellar charges banks to use the platform. Ripple is semi-decentralized because it is managed by the company and has a CEO. Ripple's market value is currently far more than Stellar. However, XRP is not the entire application of Ripple, and its application in international exchange xCurrent will be described in detail later.


To solve the broader international exchange application scenarios, scalability and throughput are an inevitable issue to discuss. A key aspect of scalability is reaching consensus on which transactions are valid and trustworthy, and when these transactions are officially pushed to the blockchain. Taking Bitcoin as an example, a transaction can be valid on the blockchain, but if the possibility of a transaction being chained under most consensus is not high, the transaction cannot be counted. As far as Bitcoin is concerned, transaction confirmation may take more than an hour to be fully determined, and this transaction speed is undoubtedly not enough for large-scale applications.

Consensus algorithms have a huge impact on the speed, scalability, and sustainability of blockchain technology. In theory, each transaction broadcast to the network requires a communication link between each node in the network. Although these systems do their best to broadcast, they are similar to having n * (n-1) / 2 communication lines. This solution is neither feasible nor scalable. With the complexity of O (n ^ 2), even if there are only 10,000 users, the network must maintain about 100 million links in order to update everyone's information in real time. However, in reality, all nodes will not maintain contact with every other node, reducing the number of necessary communication links at the cost of information transparency and time to reach consensus.

The scalability of blockchain technology can be divided into several categories, the most important of which are:

  • Transactions per second
  • Network bandwidth usage
  • data storage

The ideal blockchain technology has maximum transaction throughput, minimal network bandwidth usage, and requires as little storage as possible.

However, the specific characteristics of these technologies cancel each other out and influence each other. In a solution where all nodes are communicating and kept up to date, network bandwidth can be an issue. In addition, current public blockchain consensus protocols inherently face a major scalability limitation. Due to the distributed nature of the system, all nodes need to verify and process each transaction. Therefore, the transaction throughput per second is limited by the transaction processing capacity of each individual node in the network. As more nodes are added to the network, the distributed infrastructure also weakens, as the delay between nodes increases logarithmically as each other node increases. In addition, each node must track the entire state of the system, including the entire blockchain, which is not very sustainable in terms of scalability.

Pain points in the international exchange industry

There is a clear trade-off between decentralization and transaction processing throughput. According to Manny Trillo, senior vice president of network processing at VISA, VISA is currently capable of processing at least 56,000 transactions per second. Bitcoin blocks also have a maximum size of up to 1MB, which limits the maximum transaction throughput to approximately 7 transactions per second, which is quite far from VISA's 56,000 transactions per second. Ethereum's gas (Note 3) limit limits block size in a manner similar to Bitcoin's 1 MB limit. The difference is that Ethereum's gas limit is set dynamically by miners, while Bitcoin's block size limit is hard-coded into the protocol. With the current 8 million gas limit and an average use of 21K gas per transaction, the average number of transactions per block is slightly less than 400. A block generates up to 20 transactions per second, but this shows that more complex transactions are like smart contracts, and the more realistic number is about 10 transactions per second. In theory, distributed consensus mechanisms can provide some key benefits, such as a firm guarantee of security, fault tolerance, authenticity, and political neutrality, but at the cost of scalability. The number of transactions processed by the blockchain must not exceed the number of transactions of a single participating node in the network. The technology available today is not enough to replace the current comprehensive currency centralized system with a distributed blockchain system.

Note 3: Gas is a unit used to measure the amount of calculations required to perform certain operations.

The scalability limitations mentioned above will also restrict access to these blockchain technologies to a limited number of users with sufficient storage and computing power. If these systems do work on a global scale, the specifications required to be part of the network will go far beyond the expectations of personal computers. The elasticity of Bitcoin depends on the distribution and independence of validators. As of July 2017, six mining agencies represented the majority (nearly 60%) of hash power in the network. Therefore, when the security of the entire Bitcoin network depends on six companies, consensus cannot be reached and consensus cannot be ensured by mining capabilities alone. Another scalability issue is the significant energy consumption associated with proof of work done by certain blockchain technologies, such as Bitcoin. As mentioned in the section on Bitcoin, the difficulty of the puzzle depends only on the computing power of the miners, so collective systems are actually a good example of the tragedy of the commons. For each new CPU added to the mining pool, the difficulty must be slightly more difficult to keep the average block mining frequency at once every 10 minutes. The tremendous energy invested in the Bitcoin system has one purpose: to maintain decentralization.

Summary of exchange service providers

The average cost of a global remittance of more than $ 200 is currently around 7%. According to the World Bank, the entire market consists of $ 689 billion in total liquidity, which means that approximately $ 48 billion is transferred directly from remittances through fees, intermediaries, and financial institutions ($ 48 billion per year). In addition, 80% of total funds in the remittance market come from emerging economies.

Not only are the user costs high, but also bank accounts and intermediaries (each remittance transaction involves at least three third parties), and the transaction time is still prohibitively high. International currency exchange is a slow and expensive process if conducted through a conventional banking system. Alternatively, paying directly with a foreign visa card will result in visa's own exchange rate plus the inevitable 2.0% conversion fee. Using convenient, dedicated currency exchange and mailing services like Forex or Western Union will be more expensive as they charge around 5-10%.

The ideal way to exchange funds across currencies is to exchange them with the opposite motivation. If such a transaction is to take place, the parties will meet in person to agree on the appropriate exchange rate and exchange funds without charging any fees or transaction costs. Without proper channels to contact legal persons for transactions, we will not be able to use traditional banking services. Banks' outstanding distribution channels, reliability and availability make them the most popular services today. Banks offer various options to meet the different needs of currency exchange:

  • Direct cash exchange at bank branches
  • International bank transfer
  • Open a currency account at a local bank

Currency banks such as Western Union and Forex are the first point, and by far the most expensive alternative, they also offer plenty of currency services at convenient tourist locations such as airports and city centres. A currency account is also an expensive option and it cannot be connected to the card. Only large banks offer this option, and each transaction charges a high fee.

For people planning to stay longer in another country, it is financially reasonable to create a bank account in that country compared to using only a debit / credit card. International bank transfers can be a convenient way to fill out a new bank account. International bank transfers are one of the services that banks have recently competed against with companies like Transferwise, WeSwap and Currency Fair. These traditional solutions all offer much cheaper international currency exchange rates than banks, but their concepts are slightly different.

Traditional solutions

Transferwise Among non-bank international currency exchange services, Transferwise is probably the most popular service. Since 2011, Transferwise has made transfers with prominent investors on the team such as Richard Branson and Peter Thiel. Their concept is based on using a buffered account in each country or territory, and never actually transferring the amount, but receiving money in one currency and paying in another currency. They use the official mid-market exchange rate without any markup. When many people use the service, they can even write off each other statistically. In this way, Transferwise can eliminate transaction fees and charge a smaller fee (about 1%) than the bank for that trouble. The solution is simple and can be used as a web application.

Currency Fair Currency Fair has the same model as the stock market. One can set a bid for the currency and then wait for someone to accept it. Therefore, users can set their own exchange rate, and Currency Fair charges 0.15% of the total. If no one goes there to redeem, Currency Fair will initiate the redemption on its own, with a fixed fee of $ 3 and a hidden redemption spread of 0.9%. Therefore, Currency Fair gives customers the freedom to set a given exchange rate, wait for someone to accept it, or immediately sell it to Currency Fair at a slightly higher cost, but not the same as what banks charge.

WeSwap WeSwap provides customers with MasterCard, which they can recharge in their national currency. Clients can then conduct currency transactions internally with other WeSwap clients with opposite needs. Transaction fees vary between 1% and 2%, depending on the urgency of the transaction. This service is for users who do not want to use their primary personal card for currency exchange or users who do not have a bank account in a second currency.

SWIFT SWIFT stands for "Global Interbank Financial Telecommunications Association". It is a global network of financial organizations used to exchange financial transaction information and can be understood as the communication medium for all banks and financial institutions. All users in the network have a service identifier code or SWIFT code for communication.

The platform was established in 1973 by a Belgian company called Telecommunications. Today, most banks use this network to maintain and trade global interbank currencies. SWIFT provides a network that enables financial institutions to send and receive information about financial transactions in a secure, standardized, and reliable environment. Most banks use the SWIFT network for remittances. As of September 2010, more than 9,000 financial institutions in 209 countries worldwide sent and received more than 15 million messages per day, compared to 2.4 million in 1995. In addition, the bank's daily currency transaction volume is approximately $ 5 trillion processed separately, most of which is handled by message exchanges with the SWIFT network.

SWIFT does not understand the foreign exchange component of cross-border payments, but it is a partner of CLS, a foreign exchange settlement system. CLS's own members include many of the world's largest financial institutions and all of their major central banks.

Blockchain solution

xCurrent Ripple has created an open source product xCurrent that allows peer-to-peer financial transactions between banks worldwide. All members of RippleNet are connected through Ripple's standardized technology xCurrent. xCurrent enables banks to send messages and settle with RippleNet member banks faster, with greater transparency and efficiency. xCurrent is built around the Interledger Protocol (ILP), an open, neutral protocol that enables interoperability between different ledgers and payment networks. The solution provides cryptographically secure end-to-end payment processes with transaction invariance and information redundancy. RippleNet is not a completely decentralized technology, non-standardized communication and centralized network, but a global banking network. Banks can send and receive payments via Ripple's distributed financial technology.

Circle Circle was founded in 2013 as an original wallet. In 2015, it became the first company in the world to obtain a digital currency license (BitLicense) from the New York State Financial Services Authority (NYDFS). In April 2016, Circle obtained the first E-Money Issuer License issued by the UK's highest financial regulatory agency, the UK Financial Conduct Authority, which can be legally issued in the UK and EU countries. Carry out the exchange business of fiat and digital currencies.

In cross-border payment, Circle mainly implements low-cost exchange of currencies and cross-border exchanges with the support of blockchain technology. Users can transfer, receive and pay without commission. Circle customers already cover more than 150 countries and territories, with 70% of users concentrated in the United States and Europe. The annual transaction volume is close to US $ 1 billion, and global customers in 2016 increased by 300% year-on-year.




With changes in domestic habits and needs, and the introduction of real-time domestic payment systems, and the shift of the local real-time total settlement system to 24/7 settlement, the banking industry is experiencing a wave of automation.

In November 2018, the European Union launched the "targeted instant payment settlement" system (commonly known as TIPS); people and companies in Europe can now make instant euro payments to each other 24/7, 365 days a year. In the United States, the clearing house (TCH) is gradually rolling out 24/7/365 instant payments to individuals and companies. To survive, banks must deal with these profound changes. Of course, there is no doubt that SWIFT and its member banks must deal with the threat posed by cryptocurrencies if they are to survive.

SWIFT is not nothing. Two years ago, it shocked Ripple's instant international payment product with its Global Payments Plan (GPI). SWIFT belongs to most of the major banks in the world among its members. They inevitably adopted GPIs: within a year, 160 banks were using GPIs. In response, Ripple adopted the cryptocurrency XRP as an international payment "standard" and focused on application-based retail payments through its partner banks. But now, SWIFT is keeping up with this market. SWIFT made it clear that by 2020, GPI will become the standard for all cross-border payments.

SWIFT's proposal builds on the GPI program to create an instant, uninterrupted, and ubiquitous international payment network. SWIFT does not believe in the value of blockchain. It plans to rely on more general technologies: core architecture, common standards and APIs.

The lack of common standards is the biggest obstacle to seamless international payments. Therefore, the transition to the open ISO 20022 standard by 2021 is a prerequisite for the SWIFT program. SWIFT is passionate about the benefits of using standards:

The importance of switching to ISO 20022 cannot be underestimated. It will enable banks to direct instant payments across borders through domestic systems to the account of the ultimate beneficiary. It will allow smaller markets to internationalize and expand more quickly. It will enable larger markets to move smoothly across the entire currency “stack” within their markets, including flows between banks and other payment providers, credit card companies and local clearing houses. It will increase competition, drive innovation and reduce industry friction. In terms of speed, ubiquity and choice, it will benefit the end customer. It will also bring significant benefits to the industry in reducing total cost of ownership, simplifying integration and efficient business processes.

Ripple system

The Ripple protocol maintains a distributed ledger that is common across the entire network. Ripple's "consensus mechanism" allows all nodes in the system to automatically receive updates to the general ledger transaction records within a few seconds. This process does not need to go through a central data processing center. This rapid processing method is the most significant technological breakthrough of the Ripple system.

Compared with the data packaging method and transaction confirmation method, Ripple system has two differences: First, the block packaging speed is faster. Bitcoin is about ten minutes or so, and the Ripple system only takes a few seconds). The second is the faster block confirmation method. Bitcoin requires multiple nodes to confirm one by one, and the Ripple system is the simultaneous confirmation of all nodes together. This is the "consensus mechanism" of the Ripple system. Therefore, the confirmation time of the new transaction record of the Ripple system takes only 3-5 seconds, while Bitcoin generally takes 40-50 minutes.

The Ripple system is a decentralized architecture, while SWIFT is a centralized architecture. Centralized system maintenance costs are relatively high, such as employee wages, server and other equipment costs, and necessary profits, which all lead to a centralized system. May be "free" as required by Internet finance. Ripple's decentralized architecture does not have these problems.

At the same time, the Ripple system's transactions are lower than SWIFT. SWIFT is a centralized structure, so cross-currency, cross-national, and cross-region operating costs and charging standards are relatively high. Any currency in the Ripple system can be freely converted at almost zero cost, and cross-currency transfers are very convenient. Moreover, in the Ripple system, there is no difference in cross-region, inter-bank, and cross-border payments, and all are close to zero cost.

The settlement speed of the Ripple system is much faster than SWIFT, and the transfer of the Ripple system will arrive in 5 seconds. In contrast, SWIFT remittances take 1-2 days.

Ripple system transactions can be anonymous, while SWIFT transactions require the identification of both parties.

The Ripple system can be used in any currency, while SWIFT can only be used in national currencies. As long as there is an exchange rate, any currency can be freely converted.


By far, the most successful application of blockchain in international exchange is Ripple's xCurrent. Ripple is also the company most promising to replace SWIFT as another giant in international exchange. According to the McKinsey report, B2B cross-border payment using blockchain can add 50-50 billion US dollars in value to the global payment industry, which comes from reducing costs, accelerating speed, expanding markets, and increasing security. Unlike other functional applications, Ripple points to the essence and holds the most revolutionary application of blockchain, and the largest piece of cake in its hands. The company that masters international exchange will certainly occupy the most important place in future international trade.

risk warning:

  • Beware of illegal financial activities under the banner of blockchain and new technologies. The standard consensus resolutely resists illegal activities such as illegal fundraising, online pyramid schemes, ICOs, various variants, and dissemination of bad information.