Benchmarking the performance of a self-custody, non-ledger-based, obliviously managed digital payment system
William Macpherson, Geoffrey Goodell
TL;DR
This paper investigates a privacy-preserving retail CBDC architecture by extending the Comet framework with a Cumulative Merkle Trie to accelerate Proofs of Provenance (PoP) verification. It redesigns a private DLT around Raft consensus, omitting heavy features like smart contracts to maximize efficiency while preserving immutability and privacy, and it implements a Relay Service that aggregates transactions into a Cycle Trie before disseminating Merkle Roots to a ledger. The authors provide theoretical analyses of parameter relationships ($n$, $N$, $D$, $T_c$, $T_p$, $R_{ctp}$, $R_{pr}$, $L$) and present empirical results showing faster PoP retrieval and favorable throughput for the novel design versus a legacy approach, while candidly addressing limitations and future work. The work offers a scalable blueprint for privacy-preserving, high-throughput CBDC infrastructures capable of processing thousands of transactions in real time, with practical implications for national payment systems.
Abstract
As global governments intensify efforts to operationalize retail central bank digital currencies (CBDCs), the imperative for architectures that preserve user privacy has never been more pronounced. This paper advances an existing retail CBDC framework developed at University College London. Utilizing the capabilities of the Comet research framework, our proposed design allows users to retain direct custody of their assets without the need for intermediary service providers, all while preserving transactional anonymity. The study unveils a novel technique to expedite the retrieval of Proof of Provenance, significantly accelerating the verification of transaction legitimacy through the refinement of Merkle Trie structures. In parallel, we introduce a streamlined Digital Ledger designed to offer fast, immutable, and decentralized transaction validation within a permissioned ecosystem. The ultimate objective of this research is to benchmark the performance of the legacy system formulated by the original Comet research team against the newly devised system elucidated in this paper. Our endeavour is to establish a foundational design for a scalable national infrastructure proficient in seamlessly processing thousands of transactions in real-time, without compromising consumer privacy or data integrity.