Lightweight Blockchain-based Trade Tracking: A Threshold-Optimized Algorithm with Asynchronous Resilience, Equitable Compliance, and Green Verifiability

Abstract

Global cross-border trade, exceeding $20 trillion annually, demands blockchain traceability that simultaneously guarantees subsecond asynchronous query availability, fork-free regulatory evolution, and collusion-resistant credential confidentiality. Existing systems, however, fall into a trilemma challenge—static thresholds livelock under network jitter, hard-coded policies force disruptive hard forks, and naïve fragmentation permits credential reconstruction. We identify three deeper root causes previously overlooked: dimensional mismatch between latency values and integer quorums, conflation of erasure coding with cryptographic confidentiality, and inconsistent fault-tolerance assumptions between ordering and peer layers. This paper presents a lightweight permissioned blockchain that fundamentally resolves these disconnects through three synergistic innovations. A dimension-consistent normalized negative-feedback controller eliminates algebraic mismatches, proving that dynamic threshold adjustment can sustain O(log n) queries with zero timeouts even under 500 ms artificial delay injection. Crucially, we correct a widespread cryptographic misconception by strictly decoupling confidentiality from erasure recovery: AES-256 encryption safeguarded by Shamir threshold key sharing provides provable plaintext collusion resistance, while RaptorQ fountain codes are repurposed exclusively for ciphertext fragment loss recovery—without any native confidentiality claim. We further draw a clean threat boundary separating crash-only Raft ordering from Byzantine-prone lightweight peers, confining asynchronous resilience claims strictly to the query and signature aggregation layer. Deployed across a 100-node intercontinental testbed spanning five AWS® regions and the Greater Bay Area, our prototype demonstrates transformative results: under 30% malicious jitter, p99 latency reaches 78 ms — a six-fold improvement over static baselines—with zero query timeouts, while mean latency holds at 24 ms. The system compresses on-chain storage by 95.3%, reduces carbon footprint by 62% (saving 15,800 tonnes CO2e annually), and cuts audit costs by 67% compared to Hyperledger® Fabric. This work offers the first mathematically rigorous and experimentally validated solution to the trade traceability trilemma, delivering a verifiably green, compliance-ready blueprint for sovereignty-aware cross-border governance. All source code is open-sourced for full reproducibility.