Search for: All records

The Bitcoin scalability problem has led to the development of offchain financial mechanisms such as payment channel networks (PCNs) which help users process transactions of varying amounts, including micro-payment transactions, without writing each transaction to the blockchain. Since PCNs only allow path-based transactions, effective, secure routing protocols that find a path between a sender and receiver are fundamental to PCN operations. In this paper, we propose RACED, a routing protocol that leverages the idea of Distributed Hash Tables (DHTs) to route transactions in PCNs in a fast and secure way. Our experiments on real-world transaction datasets show that RACED gives an average transaction success ratio of 98.74%, an average pathfinding time of 31.242 seconds, which is 1.65 × 103, 1.8 × 103, and 4 × 102 times faster than three other recent routing protocols that offer comparable security/privacy properties. We rigorously analyze and prove the security of RACED in the Universal Composability framework. 

The Bitcoin blockchain scalability problem has inspired several offchain solutions for enabling cryptocurrency transactions, of which Layer-2 systems such as payment channel networks (PCNs) have emerged as a frontrunner. PCNs allow for path-based transactions between users without the need to access the blockchain. These path-based transactions are possible only if a suitable path exists from the sender of a payment to the receiver. In this paper, we propose Auroch, a distributed auction-based pathfinding and routing protocol that takes into account the routing fees charged by nodes along a path. Unlike other routing protocols proposed for PCNs, Auroch takes routing fees into consideration. Auroch maximizes the profit that can be achieved by an intermediate node at the same time minimizing the overall payment cost for the sender. 

Money laundering using cryptocurrencies has become increasingly prevalent, and global and national regulatory authorities have announced plans to implement stringent anti-money laundering regulations. In this paper, we examine current anti-money laundering (AML) mechanisms in cryptocurrencies and payment networks from a technical and policy perspective, and point out practical challenges in implementing and enforcing them. We first discuss blacklisting, a recently proposed technique to combat money laundering, which seems appealing, but leaves several unanswered questions and challenges with regard to its enforcement. We then discuss payment networks and find that there are unique problems in the payment network domain that might require custom-designed AML solutions, as opposed to general cryptocurrency AML techniques. Finally, we examine the regulatory guidelines and recommendations as laid out by the global Financial Action Task Force (FATF), and the U.S. based Financial Crimes Enforcement Network (FinCEN), and find that there are several ambiguities in their interpretation and implementation. To quantify the effects of money laundering, we conduct experiments on real-world transaction datasets. Our goal in this paper is to survey the landscape of existing AML mechanisms, and focus the attention of the research community on this issue. Our findings indicate the community must endeavor to treat AML regulations and technical methods as an integral part of the systems they build and must strive to design solutions from the ground up that respect AML regulatory frameworks. We hope that this paper will serve as a point of reference for researchers that wish to build systems with AML mechanisms, and will help them understand the challenges that lie ahead.