Bitcoin is by far the most popular unregulated, decentralized cryptocurrency. Since the genesis block was mined, the number of bitcoin users has been rising at a rather exponential rate. While bitcoin’s public ledger offers high levels of transparency, it does not directly reveal the real-world identity of either the senders or receivers of bitcoin transactions. A single bitcoin transaction is represented by a set of inputs that point back to outputs, or the proceeding transactions, and a set of outputs, each allocating a certain amount of bitcoins that has been sent to a recipient’s bitcoin address. A bitcoin address is represented by a string of alphanumeric characters which are generated using the public key of a matching key pair created by some bitcoin user. Every user can have numerous key pairs, and bitcoin addresses, using his bitcoin wallet, and it is highly recommended to use a new bitcoin address for each and every transaction to heighten the level of anonymity of transactions.
A lot can be learned by studying the bitcoin address graphs, the scale of adoption, bitcoin remuneration, using bitcoin for storing value and others. A recently published study presented a longitudinal study that has included an analysis of the bitcoin address graph, which is comprised of all transactions and addresses right from the start of the bitcoin protocol in January, 2009, all the way to August 31, 2016. The results of this study are quite interesting. The analysis concluded that the bitcoin address graph exhibits a highly skewed degree of distribution, that is characterized by a small percentage of outliers and proves that the entire graph is expanding at an accelerating rate. Moreover, the study highlighted the power of bitcoin address clustering heuristics as means for tracing users in the real world, especially those who use bitcoin for transferring value rather than storing it.
The Goal and Framework of the Study:
The study aimed at presenting initial results derived from analyzing the bitcoin address graph longitudinally. The basic framework of the study can be described as follows:
1. The study provided a comprehensive representation of the graph that included all executed bitcoin transactions and created addresses starting from March 1, 2009 all the way down to the time of writing of the paper on October 31, 2016.
2. A structural analysis of the bitcoin address graph was meticulously conducted to investigate the structural changes of the graph over time.
3. The study investigated the proportion of bitcoin addresses that can be traced back to its actors in the real world, whether implicitly or explicitly, and showed how these patterns can change over time.
4. The study also closely examined the spending behavior of users who consider bitcoin’s real time fiat exchange rates.
5. The study analyzed the function of cryptocurrencies from users’ perspectives via thorough analysis of the periods of activity of bitcoin addresses and clusters of bitcoin addresses.
The Bitcoin Address Graph:
The below chart represents the bitcoin address graph that was included in the published paper and analyzes the evolution of number of bitcoin addresses over time.
The chart represents the overall number of bitcoin addresses, or distinct nodes, in addition to the number of newly added nodes and edges, or transactions, along with the average node degree per month. By examining the chart, you can easily discover that the number of used bitcoin addresses is steadily rising, which can be interpreted in one of two ways:
1. Users are creating new bitcoin addresses for each transaction they execute
2. New users are joining the network and creating new addresses
In reality, both scenarios are taking place simultaneously. A single user might be using multiple addresses to create lots of transactions within a given observation period. The number of transactions, or edges, is also steadily rising in the same manner, which reflects, at least partly, a steady rise in the number of users.
Implications of the study:
The analysis showed a highly skewed degree of distribution, which denotes that the bitcoin address graph is comprised mainly of a rather small proportion of outliers. Thorough examination of these bitcoin addresses revealed that most of these addresses belong to charity organizations that use them for collection of donations, and a number of online casinos. The study also showed that the bitcoin address graph is growing at a rapid rate as time goes by, as new transactions and bitcoin addresses are added everyday to the blockchain. However, the average node degree remains somehow stable over time.
Investigating real world actors has revealed that bitcoin address clustering via well known heuristics’ approaches can boost the number of implicitly recognized addresses across the entire address graph. Nevertheless, most of the addresses will remain anonymous.
The study also spotted a rise in the volume of transactions and how real world events can influence the bitcoin exchange rates. The study proved that real world actors prefer to use bitcoin for transferring value rather than storing it.