Work

Blockchain Technology for Interbank Settlement: A Permissioned Architecture Inspired by Bitcoin, with Simulation-Based Performance Evaluation

Year: 2026
Type: preprint
Abstract: Distributed ledger technology, pioneered by Bitcoin, offers a decentralized alternative to the centralized clearing hubs that currently underlie most interbank and interbranch settlement systems. This paper positions itself explicitly against prior blockchain-for-banking proposals, systematizes the core technical components of the Bitcoin architecture—ECDSA signatures, Merkle-tree block structure, Proof-of-Work (PoW) consensus, and peer-to-peer networking—and proposes a permissioned blockchain architecture for interbranch settlement, with pre-identified branches as nodes, low-calibrated PoW difficulty, Simplified Payment Verification (SPV) endpoints, and a structured STRIDE threat-model analysis. The proposal is validated through a reproducible, statistically characterized Python simulation across five experiments. Results show mining time grows approximately exponentially with PoW difficulty (log-linear R² = 0.999 on difficulty-level means; R² = 0.81 fitted directly to raw trials, with a bootstrap CI used where small, right-skewed samples make the normal-theory interval unreliable); increasing block size is the dominant throughput lever; Merkle-proof length scales as ⌈log₂ n⌉, consistent with SPV's O(log n) complexity against full-node O(n) scaling (R² = 0.99 vs. 0.79); hash-chaining with digital signatures gives immediate, verifiable tamper-evidence; and a Monte-Carlo network-latency model shows settlement finality is dominated by network propagation rather than PoW sealing at low difficulty. Measured throughput is benchmarked qualitatively against Bitcoin (≈ 7 tx/s) and Hyperledger Fabric (> 3,500 tx/s). These findings guide the calibration of PoW difficulty, block size, quorum size, and verification architecture for blockchain-based interbranch reconciliation in commercial banks.
Language: English
Open Access status: green
OpenAlex ID: W7236427054
Cites: 23
Cited by:
Related to: 10