|
MoneroResearch.info |
 |
 |
 |
|
Ni, W., Cheng, P., Chen, L., & Lin, X. (2021). When the recursive diversity anonymity meets the ring signature. In Proceedings of the 2021 International Conference on Management of Data (pp. 1359–1371). New York, NY, USA: Association for Computing Machinery. Added by: Rucknium (2022-02-23 15:57) |
| Resource type: Book Article ID no. (ISBN etc.): 9781450383431 BibTeX citation key: Ni2021 View all bibliographic details  |
Categories: Monero-focused Creators: Chen, Cheng, Lin, Ni Publisher: Association for Computing Machinery (New York, NY, USA) Collection: Proceedings of the 2021 International Conference on Management of Data |
Views: 42/1616
|
Attachments
ni2021.pdf  [17/881]
|
URLs https://doi.org/10.1145/3448016.3452825 |
| Abstract |
|
In privacy-preserving blockchain systems, to protect a sender's identity of a transaction in privacy-preserving blockchain systems, ring signature (RS) schemes have been widely implemented, which allow users to obscure consumed tokens via including "mixin'' (i.e., chaff tokens). However, recent works point out that existing RS schemes are vulnerable to the "chain-reaction'' analysis, where adversaries eliminate mixins of RSs by utilizing the fact that each token can only be consumed in a RS. By "chain-reaction'' analysis, adversaries can find some definite token-RS pair sets (DTRSs) to confirm the sender's identity of a RS. Besides, the existing RS schemes do not consider the diversity of mixins when generating a RS. Moreover, since the transaction fee is proportional to the number of mixins, a use is motivated to use a RS with the minimum number of mixins. In this paper, we formally define the diversity-aware mixins selection (DA-MS) problem, which aims to generate a RS with the minimum number of mixins satisfying the constraints of its diversity and the anonymity of other RSs. We prove the DA-MS problem is $#P$ and propose a breadth-first search algorithm to get the optimal solution. Furthermore, to efficiently solve the DA-MS problem, we propose two practical configurations and two approximation algorithms with theoretic guarantees. Through comprehensive experiments on real data sets as well as synthetic data sets, we illustrate the effectiveness and the efficiency of our solutions.
Added by: Rucknium |