Buffer-aided relay selection serves as one of the key approaches for achieving physical layer security (PLS) of wireless networks. Existing studies on buffer-aided relay selection mainly concentrate on the scenarios where transmitted packets have unlimited lifetime, while the more practical and important scenarios with limited packet lifetime remain largely unexplored. This paper therefore aims to study the buffer-aided relay selection for achieving PLS in two-hop wireless networks with limited packet lifetime. We first propose a novel buffer-aided relay selection scheme by carefully considering the remaining lifetimes of packets, conditions of available links and channel state information (CSI) of eavesdroppers. To model the security and reliability performances achieved by the proposed scheme, we then apply tools from the Markov chain theory to derive expressions of the secrecy outage probability (SOP) and the packet discarding probability (PDP) of the network, respectively. Finally, extensive simulation and numerical results are provided to illustrate the performances of our proposed scheme and our theoretical framework. The results show that although our scheme achieves a slightly higher SOP than the typical buffer-aided max-link relay selection scheme, it can greatly reduce the PDP of the network.

缓冲区辅助中继选择是实现无线网络物理层安全 (PLS) 的关键方法之一。现有关于缓冲辅助中继选择的研究主要集中在传输的数据包生命周期无限的场景，而更实用和更重要的数据包生命周期有限的场景仍未得到探索。因此，本文旨在研究缓冲辅助中继选择，以在具有有限数据包寿命的两跳无线网络中实现 PLS。我们首先通过仔细考虑数据包的剩余生命周期、可用链路的条件和窃听者的信道状态信息 (CSI) 来提出一种新的缓冲区辅助中继选择方案。为了对所提出的方案实现的安全性和可靠性性能进行建模，然后，我们应用马尔可夫链理论中的工具分别推导出网络的保密中断概率 (SOP) 和数据包丢弃概率 (PDP) 的表达式。最后，提供了广泛的模拟和数值结果来说明我们提出的方案和我们的理论框架的性能。结果表明，虽然我们的方案比典型的缓冲辅助最大链路中继选择方案实现了略高的 SOP，但它可以大大降低网络的 PDP。