Enabling Scalable and Unlinkable Payment Channel Hubs with Oblivious Puzzle Transfer
Payment channel networks (PCNs) are effective techniques for extending the scalability of cryptocurrencies. It achieves this by establishing a direct off-line channel from the sender to the receiver, going through one intermediary (aka. the hubs). In such scenarios, the hubs know the origin and destination of each transaction flowing through them, which jeopardizes the privacy of the underlying systems. Unfortunately, former efforts in ensuring transaction unlinkability either rely on trusted mixing services, are inefficient constructed (e.g., constructed inefficient cryptographic primitives), or have limited applicability. In this paper, we present ObliHub, an efficient payment channel scheme that conceals transaction direction information to the hubs. The core technique of ObliHub in achieving unlinkability is our tailored oblivious puzzle transfer protocol (OPT), which enables puzzle solving among the payer, the hub, and the receiver to be conducted in an oblivious manner – the hub center neither knows where a puzzle hint came from nor who acquired it. The implementation of ObliHub only requires efficient cryptographic primitives, and compared with
(a state-of-the-art Bitcoin-compatible PCH using homomorphic encryption), ObliHub saves 0.2 seconds in computation time over previous solutions and improves transfer throughput. Besides, our scheme is in accord with Universal Composability (UC) framework and we provide a comprehensive security analysis of it.
Foundation of National Natural Science Foundation of China (Grant No.: 62072273, 72111530206, 61962009); The Major Basic Research Project of Natural Science Foundation of Shandong Province of China (ZR2019ZD10); Natural Science Foundation of Shandong Province (ZR2019MF062); Shandong University of Science and Technology Program Project (J18A326); Guangxi Key Laboratory of Cryptography and Information Security (No.: GCIS202112); Foundation of Guizhou Provincial Key Laboratory of Public Big Data (No. 2019BD-KFJJ009); This work was supported by the Key-Area Research and Development Program of Guangdong Province (No. 2020B0101130015)
Author affiliationSchool of Computing and Mathematical Sciences, University of Leicester
- AM (Accepted Manuscript)