On the suitability of parallel computer architectures for attacks on lattice-based cryptography

随着量子计算机的发展，密码学的前景将不得不重新定义，因为目前的密码系统将容易受到量子攻击。目前，基于格子的密码系统是后量子计算非常有前途的候选者。然而，我们对其安全性的理解仍在不断成熟。特别是，我们需要评估针对这些密码系统的攻击的潜力，以便我们能够确定它们的实际安全性。这也有助于我们确定密码系统的安全参数，因为它们是根据攻击的实际性能在强大的并行计算机架构上设置的。在这个项目中，我们建议1)加深我们对基于格的密码系统攻击的了解，为此，我们将使用不同的计算机体系结构；2)开发工具，用于开发实用的、并行的基于格的密码分析。

With the development of quantum computers, the landscape of cryptography will have to be redefined, given that current cryptosystems will be vulnerable against quantum attacks. Currently, lattice-based cryptosystems are very promising candidates for postquantum computing. However, we are still maturing our understanding of their security. In particular, we need to assess the potential of attacks against these cryptosystems, so we are able to determine their actual safeness. This also helps us to determine the security parameters for the cryptosystems, as they are set based on the practical performance of attacks, on powerful, parallel computer architectures. In this project, we propose to 1) deepen our knowledge of attacks against latticebased cryptosystems, to which end we will use different computer architectures and 2) develop tools for the development of practical, parallel lattice-based cryptanalysis.