[1]ZAI Guangjun,DONG Yannan,WANG Yipeng,et al.Anti-Correlation Two-party Blockchain Covert Communication Method Based on Dual-Channel Dynamic Evolution[J].Journal of Zhengzhou University (Engineering Science),2027,48(XX):1-9.[doi:10.13705/j.issn.1671-6833.2026.06.016]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
48
Number of periods:
2027 XX
Page number:
1-9
Column:
Public date:
2027-12-10
- Title:
-
Anti-Correlation Two-party Blockchain Covert Communication Method Based on Dual-Channel Dynamic Evolution
- Author(s):
-
ZAI Guangjun,DONG Yannan,WANG Yipeng,XU Zhenyu,SHE Wei
-
School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450001, China
-
- Keywords:
-
covert communication; blockchain; zero-knowledge proof; forward secrecy; anti-correlation analysis
- CLC:
-
TP393.0;TN915.08
- DOI:
-
10.13705/j.issn.1671-6833.2026.06.016
- Abstract:
-
To address the core problems of lacking forward secrecy, weak anti-correlation analysis capabilities, and incomplete communication processes universally existing in current blockchain covert communication schemes, a self-evolving closed-loop covert communication protocol (SECL-CCP) was proposed. Through three innovative mechanisms, a security architecture capable of dynamic evolution across keys, protocol processes, and on-chain infrastructure was constructed. First, a "key ratchet" mechanism was designed, wherein the "commit-reveal" scheme was combined with the public entropy of blockchain future states, irreversible key updates were realized, and forward secrecy was endowed to the protocol. Secondly, a "dynamic contract factory" model was proposed, and the CREATE2 opcode was utilized to dynamically deploy disposable interactive contracts for each communication round, whereby fixed on-chain fingerprints were eliminated. Finally, a receipt mechanism based on zero-knowledge proof was introduced to build a verifiable communication closed loop, and the receiver non-repudiation problem was resolved. Simulation experiments were conducted based on the Ethereum Sepolia testnet and Hardhat environment, and a mixed dataset containing 20,000 samples was constructed by collecting real mainnet traffic. It was demonstrated by the results that the steganalysis detection rates under random forest and LSTM models were respectively reduced to 12.8% and 11.5%, zero leakage of historical information under key compromise scenarios was achieved, and intelligent correlation analysis attacks were effectively resisted