TL;DR
This paper introduces SecurePay, a novel payment system for the platform economy that combines permissioned blockchains with Central Bank Digital Currencies (CBDCs) to achieve both robust fund and information security without sacrificing processing speed. By leveraging smart contracts and a multi-party endorsement system, SecurePay aims to overcome the trust dilemmas inherent in current e-commerce payment systems.
What Problem Does This Paper Address?
The rapid growth of the platform economy, while offering convenience, faces significant security challenges, particularly concerning payment processing. E-commerce platforms often act as trusted intermediaries, holding funds in escrow. This centralized model creates a "trust dilemma" where malicious platforms can misuse user funds and sensitive transaction data. Existing solutions either prioritize security at the cost of performance or vice versa, failing to adequately address both fund security (preventing misuse during the escrow period) and information integrity (preventing unauthorized alteration of transaction details). Furthermore, current systems often compromise user privacy by requiring extensive data sharing.
Methods
SecurePay employs a hybrid on-chain and off-chain architecture.
- Permissioned Blockchain: A permissioned blockchain is used for processing transaction information and deploying smart contracts. This allows for controlled participation and ensures that only authorized entities (banks, merchants, e-commerce platforms) can endorse transactions. A multi-party endorsement protocol is central to verifying transaction integrity.
- Central Bank Digital Currency (CBDC): CBDCs are utilized for managing payments. Their programmability via smart contracts enables the creation of secure escrow wallets that automatically lock and release funds based on pre-defined rules. This feature is crucial for ensuring fund security and preventing misuse during the escrow period. CBDCs also offer inherent privacy-preserving features.
- On-chain/Off-chain Separation: To enhance performance and user privacy, payment processing is separated from on-chain activities. CBDC payments are verified off-chain by banks, preventing the exposure of full wallet balances to unauthorized parties. This also mitigates the volatility risks associated with purely on-chain cryptocurrencies.
Datasets
Not discussed in the paper. The paper focuses on a system implementation and experimental evaluation rather than training on specific datasets.
Key Insights and Ideas
- CBDC as a Game Changer for Escrow: The programmability of CBDCs is a critical enabler for secure escrow. It allows for automated fund locking and release based on verifiable on-chain events, directly addressing the vulnerability of funds being held by a central intermediary.
- Multi-Party Endorsement for Information Integrity: Moving beyond a single platform's control, requiring endorsements from multiple stakeholders (banks, merchants, platform) significantly strengthens the integrity of transaction information and makes collusion harder.
- Hybrid Architecture Balances Security and Performance: The separation of on-chain (information processing, endorsement) and off-chain (payment settlement) activities is key to achieving both high security and competitive transaction speeds, a common trade-off in blockchain systems.
- User Privacy through Controlled Data Access: By using CBDCs for off-chain verification and limiting on-chain data visibility, SecurePay aims to protect user privacy while still allowing for necessary transaction validation.
Key Results
The paper reports experimental results from a deployed SecurePay system:
- Normal Conditions: Achieved a transaction throughput of 256.4 transactions per second (TPS) with an average latency of 4.29 seconds. This performance is comparable to centralized systems and significantly outperforms existing decentralized solutions in terms of latency.
- Reshipping Attack Simulation: Under simulated reshipping attack conditions (where a buyer falsely claims non-receipt), the system's TPS dropped to 86.4 TPS with an increased average latency of 4.81 seconds.
- Settlement Inconsistency Simulation: When simulating settlement inconsistency (requiring banks to provide cryptographic proofs), TPS decreased to 81.4 TPS with an average latency of 5.07 seconds.
Limitations and Future Work
- Reshipping Attack and Settlement Inconsistency: While the system has mechanisms to handle these, performance is impacted. Future work could focus on optimizing these specific workflows.
- Bottleneck in Multi-Party Endorsement: The paper identifies the multi-party endorsement phase as a bottleneck for throughput. Future enhancements will focus on optimizing smart contract logic, including batch signatures and batch payment information updates.
- Scalability: While experiments were conducted on personal devices and servers, further large-scale testing would be beneficial.
- Adoption and Interoperability: The paper doesn't deeply explore the challenges of widespread adoption or interoperability with existing financial systems beyond the CBDC integration.
Summary updated