Abstract
As blockchain ecosystems grow, the need for secure cross-chain asset transfers becomes critical. Current notary-based mechanisms often suffer from collusion risks, biased credit evaluations, and barriers to entry for new nodes. This paper proposes a novel cross-chain asset trading scheme utilizing edge cloud storage and a multi-indicator credit evaluation system. By integrating the Entropy Weight Method (EWM) with a PageRank-inspired ranking algorithm, the scheme ensures objective node assessment and fair resource allocation. Experimental results demonstrate a significant reduction in error rates and enhanced resilience against malicious nodes compared to existing models.
Introduction
The expansion of blockchain technology has created a fragmented landscape of "information silos." Interoperability between disparate chains, such as Ethereum and Hyperledger, is hindered by differing consensus mechanisms and data structures. Notary mechanisms—trusted third parties that verify transactions across chains—are commonly employed to bridge this gap. However, these systems face significant challenges:
- Collusion: Notaries may conspire with users to validate fraudulent transactions.
- Inaccurate Credit Evaluation: Reliance on single-metric reliability scores makes systems vulnerable to manipulation.
- Centralization: Existing systems often favor established "veteran" nodes, preventing honest newcomers from participating and leading to the "Old Guard" problem.
Proposed Methodology
The proposed scheme introduces a multi-faceted approach to enhance notary reliability and system fairness.
Multi-Indicator Credit Evaluation
Rather than relying on a single metric, the system evaluates nodes based on a comprehensive set of indicators: user feedback, transaction success rates, processing efficiency, and negative reports. The Entropy Weight Method (EWM) is applied to these indicators to objectively assign weights, ensuring that no single metric can be easily gamed to inflate a node's reputation.
PageRank-Inspired Credit Ranking
The system treats notary nodes as a network, similar to Google's PageRank algorithm. A node's credit is a composite of its intrinsic value (performance metrics) and indirect trust (vouching from other high-credit nodes). A damping coefficient is incorporated to account for temporal decay, ensuring that past performance does not provide a permanent shield for nodes that begin to exhibit poor behavior.
Edge Cloud Storage and Management
Historical transaction data and trust relationship tables are offloaded to edge cloud servers. This architecture provides low-latency access for real-time credit calculations while maintaining decentralization.
Dynamic Node Allocation
To combat centralization, the paper introduces a dynamic share allocation function, $F(X)$. This mathematical model determines the optimal balance between "new" and "old" nodes, allowing newcomers to enter the pool without compromising the overall transaction success rate. Furthermore, a Margin Pool (security deposit) acts as a regulator; malicious behavior results in "slashing," where deposits are used to compensate for losses.
Experimental Results
The scheme was evaluated through controlled simulations involving 100 established nodes and 50 new nodes. Performance was benchmarked against the "Dai" and "Jiang" schemes across varying transaction volumes (30 to 960 transactions).
- Accuracy: The proposed scheme achieved an error rate of only 0.6% in identifying the top and bottom 10% of nodes, significantly lower than the 4.3% observed in competing models.
- Resilience: The system maintained high success rates even when the proportion of malicious nodes reached 40%, whereas the Dai scheme's performance plummeted.
- Fairness: Unlike the Dai scheme, which failed to integrate any new nodes, the proposed model successfully incorporated 25 new nodes into the notary group while maintaining system stability.
Conclusion
By shifting from subjective to objective evaluation and treating trust as a networked property, this scheme provides a robust framework for cross-chain asset trading. Future work will focus on refining economic incentive mechanisms beyond the deposit pool and addressing potential privacy concerns related to the storage of historical transaction data on edge cloud servers.
Summary updated