Traditional security models for Nakamoto-style blockchains assume instantaneous synchronization among malicious nodes, which overestimate adversarial coordination capability. We revisit these existing models and propose two more realistic security models. First, we propose the static delay model. This model first incorporates adversarial communication delay. It quantifies how the delay constrains the effective growth rate of private chains and yields a closed-form expression for the security threshold. Second, we propose the dynamic delay model that further captures the decay of adversarial corruption capability and the total adversarial delay window. Theoretical analysis shows that private attacks remain optimal under both models. Finally, we prove that large-scale Nakamoto-style blockchains offer better security. This result provided a theoretical foundation for optimizing consensus protocols and assessing the robustness of large-scale blockchains.

翻译：传统Nakamoto式区块链安全模型假设恶意节点间可实现瞬时同步，这高估了攻击者的协调能力。我们重新审视现有模型，提出了两种更贴近现实的安全模型。首先，我们提出静态延迟模型。该模型首次纳入攻击者通信延迟因素，量化了延迟如何制约私有链的有效增长率，并推导出安全阈值的闭式表达式。其次，我们提出动态延迟模型，进一步刻画攻击者腐化能力的衰减过程及总攻击延迟窗口。理论分析表明，在这两种模型下私有攻击仍保持最优性。最后，我们证明大规模Nakamoto式区块链具有更优越的安全性。该结果为优化共识协议及评估大规模区块链的鲁棒性提供了理论基础。