Privacy-Protected Deletable Blockchain

Xianji Cai, Yanli Ren, Xinpeng ZhangIEEE Access 2020 · DOI

Introduction

Blockchain technology is renowned for its immutability, a feature that ensures data integrity but creates significant challenges for regulatory compliance. Specifically, the "right to be forgotten" under the General Data Protection Regulation (GDPR) and the need to remove illegal content conflict with the permanent nature of traditional ledgers. Current solutions for "redactable" blockchains often rely on centralized authorities or require users to sacrifice anonymity to justify deletions. This paper proposes a decentralized, privacy-preserving deletable blockchain that leverages Proof-of-Space (PoSpace) consensus and a novel Linkable Digital Multi-Signature (LDMS) scheme to resolve this "privacy paradox."

Methodology

The proposed system rests on three technological pillars designed to balance accountability with anonymity.

Proof-of-Space (PoSpace) Consensus

Unlike energy-intensive Proof-of-Work, this system utilizes disk space for consensus. This provides a more resource-efficient foundation for both mining and the subsequent deletion simulations.

Privacy-Preserving Protocols

To ensure transaction privacy, the researchers integrated Ring Confidential Transactions (Ring CT). This technology hides the origins, destinations, and amounts of transactions using stealth addresses and Pedersen commitments, ensuring that the public ledger does not leak sensitive user data.

Linkable Digital Multi-Signature (LDMS)

The core innovation is the LDMS scheme. When a deletion is requested, a group of users must reach a consensus. They sign the request using one-time pseudonyms. The "linkable" property ensures security: if a malicious actor attempts to vote multiple times using different pseudonyms, the algorithm mathematically links the signatures, exposing the fraud.

The dual-chain structure of the proposed blockchain, showing the parallel proof and signature chains that maintain integrity during data removal.

Deletion Logic and Accountability

The system employs "conditional disclosure" based on the reason for deletion. If a block is targeted for deletion due to an invalid identity, the requester must reveal their identity via a traceable ring signature. If the deletion is due to expired or illegal content, the specific transaction content is revealed via a Pedersen commitment. This ensures that only the necessary evidence is disclosed to the community for verification, maintaining maximum privacy for all other transaction details.

Experimental Evaluation

The authors evaluated the protocol using a custom simulation environment written in C++. The setup utilized the TomMath library for big integer arithmetic and ECC-200 for elliptic curve operations.

A tree-based graph structure representing the hierarchical organization used in the simulation of the consensus mechanism.

The simulation involved a network of 7 miners with varying storage capacities. Key performance metrics indicated that while deletion is more computationally intensive than block generation, it remains feasible for real-world applications:

  • Average Block Generation Time: 1740ms
  • Average Block Deletion Time: 3556ms

The increased time for deletion is attributed to the complex cryptographic verification required to maintain privacy and security during the multi-signature process.

Structural Integrity

A critical challenge in deletable blockchains is maintaining the "hash chain" once data is removed. The proposed system solves this by replacing deleted transaction data with the multi-signature of the verifying voters. This allows the blockchain to maintain its structural integrity and cryptographic links without retaining the prohibited data.

Conclusion and Future Work

This research demonstrates that decentralized data deletion is possible without compromising user privacy. However, challenges remain regarding scalability. While the mathematical proofs hold, the latency of a 2/3 majority voting system in a massive, global network requires further investigation. Future work will also need to address the secure storage of the "evidence" used to justify deletions to prevent secondary privacy leaks.

Summary updated

BibTeX
@article{cai2020privacy,
  author = {Xianji Cai and Yanli Ren and Xinpeng Zhang},
  title = {{Privacy-Protected Deletable Blockchain}},
  journal = {IEEE Access},
  year = {2020},
  doi = {10.1109/access.2019.2962816},
}
Figure