Paper 2022/123

CryptoMaze: Privacy-Preserving Splitting of Off-Chain Payments

Subhra Mazumdar and Sushmita Ruj


Payment Channel Networks or PCNs solve the problem of scalability in Blockchain by executing payments off-chain. Due to a lack of sufficient capacity in the network, high-valued payments are split and routed via multiple paths. Existing multi-path payment protocols either fail to achieve atomicity or are susceptible to wormhole attack. We propose a secure and privacy-preserving atomic multi-path payment protocol CryptoMaze. Our protocol avoids the formation of multiple off-chain contracts on edges shared by the paths routing partial payments. It also guarantees unlinkability between partial payments. We provide a formal definition of the protocol in the Universal Composability framework and analyze the security. We implement CryptoMaze on several instances of Lightning Network and simulated networks. Our protocol requires 11s for routing a payment of 0.04 BTC on a network instance comprising 25600 nodes. The communication cost is less than 1MB in the worst-case. On comparing the performance of CryptoMaze with several state-of-the-art payment protocols, we observed that our protocol outperforms the rest in terms of computational cost and has a feasible communication overhead.

Note: This has been accepted for publication in IEEE Transactions on Dependable and Secure Computing (TDSC)

Available format(s)
Cryptographic protocols
Publication info
Published elsewhere. IEEE Transactions on Dependable and Secure Computing
payment channelspayment networksBitcoinwormhole attackatomic multi-path payment
Contact author(s)
subhra mazumdar1993 @ gmail com
2022-02-09: received
Short URL
Creative Commons Attribution


      author = {Subhra Mazumdar and Sushmita Ruj},
      title = {CryptoMaze: Privacy-Preserving Splitting of Off-Chain Payments},
      howpublished = {Cryptology ePrint Archive, Paper 2022/123},
      year = {2022},
      doi = {10.1109/TDSC.2022.3148476},
      note = {\url{}},
      url = {}
Note: In order to protect the privacy of readers, does not use cookies or embedded third party content.