Year | 2015 |
---|---|
Author | Patrick McCorry, Siamak F. Shahandashti, Dylan Clarke and Feng Hao |
Publisher | School of Computing Science, Newcastle University UK |
Link | View Research Paper |
Categories |
Bitcoin / Cryptocurrencies / Technical |
Bitcoin is designed to protect user anonymity (or pseudonymity) in a financial transaction, and has been increasingly adopted by major ecommerce websites such as Dell, PayPal and Expedia. While the anonymity of Bitcoin transactions has been extensively studied, little attention has been paid to the security of post-transaction correspondence. In a commercial application, the merchant and the user often need to engage in follow-up correspondence after a Bitcoin transaction is completed, e.g., to acknowledge the receipt of payment, to confirm the billing address, to arrange the product delivery, to discuss refund and so on. Currently, such follow-up correspondence is typically done in plaintext via email with no guarantee on confidentiality. Obviously, leakage of sensitive data from the correspondence (e.g., billing address) can trivially compromise the anonymity of Bitcoin users. In this paper, we initiate the first study on how to realise end-to-end secure communication between Bitcoin users in a post-transaction scenario without requiring any trusted third party or additional authentication credentials. This is an important new area that has not been covered by any IEEE or ISO/IEC security standard, as none of the existing PKI-based or password-based AKE schemes are suitable for the purpose. Instead, our idea is to leverage the Bitcoin’s append-only ledger as an additional layer of authentication between previously confirmed transactions. This naturally leads to a new category of AKE protocols that bootstrap trust entirely from the block chain. We call this new category “Bitcoin-based AKE” and present two concrete protocols: one is non-interactive with no forward secrecy, while the other is interactive with additional guarantee of forward secrecy. Finally, we present proof-of-concept prototypes for both protocols with experimental results to demonstrate their practical feasibility.