Compared to traditional hardware development methodologies, high-level synthesis (HLS) offers a faster time-to-market and lower design-cost at the expense of implementation efficiency. Although HLS tools are becoming popular in some applications, such as digital signal processing and neural network classification, their usability on cryptographic applications is largely unexplored. This feasibility is critical especially for cryptosystems that are under development, such as the next-generation public-key cryptosystems needed for quantum-resistance. This letter provides a thorough investigation of HLS on number theoretic transform (NTT)—the core arithmetic function of lattice-based quantum-resistant cryptosystems. We demonstrate a fast yet extensive design space exploration of NTT through the Vivado HLS tool, analyze the shortcomings/challenges of optimized configurations, and quantitatively compare the results to software-based and hand-coded hardware designs.

中文翻译：

---

数论变换的高级综合：未来密码系统的案例研究

与传统的硬件开发方法相比，高级综合 (HLS) 以牺牲实现效率为代价提供了更快的上市时间和更低的设计成本。尽管 HLS 工具在某些应用程序中越来越流行，例如数字信号处理和神经网络分类，但它们在密码应用程序中的可用性在很大程度上尚未得到探索。这种可行性对于正在开发的密码系统尤其重要，例如量子抗性所需的下一代公钥密码系统。这封信对数论变换 (NTT) 上的 HLS 进行了深入研究，NTT 是基于格的抗量子密码系统的核心算术函数。我们通过 Vivado HLS 工具展示了对 NTT 的快速而广泛的设计空间探索，