Achieving privacy-preserving properties in blockchain-enabled decentralized applications
Jingchi Zhang · 2025
Decentralized applications (DApps) offer significant advancements in addressing trust, privacy, and security challenges inherent in traditional centralized systems, which often suffer from single points of failure and susceptibility to censorship. While the decentralized nature of blockchain technology inherently manages trust concerns and prioritizes security, privacy has historically been treated as a secondary consideration. This thesis discusses privacy issues within DApps, specifically focusing on cloud computing and cross-chain marketplaces, and introduces enhanced schemes to safeguard user privacy without compromising system security. In the realm of cloud computing DApps, this research first identifies vulnerabilities in some of the existing Multi-authority Attribute-Based Encryption (MA-ABE) schemes with policy-hiding features, which are susceptible to rogue-key attacks under the adaptive corruption of authorities. Addressing this issue, an improved MA-ABE scheme is introduced that ensures provable security and preserves user privacy while maintaining efficiency for real-world scenarios. Building on these findings, a blockchain-based data governance system is proposed, combining our enhanced MA-ABE scheme and blockchain technology to prevent unauthorized access and privacy leakage. Leveraging smart contracts for access control and the InterPlanetary File System (IPFS) for decentralized data storage, this solution delivers a secure, decentralized, and privacy-preserving alternative to existing frameworks. For cross-chain marketplaces, this thesis introduces the Blockchain I/O framework, which in turn builds on another contribution of this thesis, namely the PIEChain sub-system -- an Apache Kafka-based approach for secure and efficient cross-chain communication -- to meet comprehensive commerce requirements. By incorporating native stablecoins to reduce price volatility and a decentralized identity (DID) system to allow privacy-preserving identity verification, the framework enables secure and private interactions across multiple blockchain networks. A proof-of-concept cross-chain marketplace confirms the feasibility of this approach, demonstrating its capacity to support efficient, privacy-centric transactions. In summary, the thesis contributes to the integration of privacy-preserving technologies into decentralized applications, offering promising solutions for both cloud computing and cross-chain marketplaces. The proposed systems not only enhance user privacy but also ensure system security and operational efficiency, thereby addressing critical gaps in existing frameworks. By meeting real-world application requirements, these contributions advance the development of DApps that prioritize both system security and user privacy.