In this paper, directional modulation (DM) is investigated to enhance physical layer security. Practical transmitter designs are exploited under imperfect channel state information (CSI) and hardware constraints, such as finite-resolution phase shifters (PSs) and per-antenna power budget. Tailored for the practical issues in realizing DM, a series of practical scenarios are investigated. Starting from the scenario where eavesdroppers (Eve)s’ information is completely unknown, corresponding designs are proposed to optimize legitimate users (LU)s’ receiving performance while randomizing the Eves’ received signal. When the Eves’ CSI is imperfectly known, in the second scenario, the Eves’ receiving performance is further deteriorated by imposing destructive interference to the Eves. For each scenario, three algorithms are proposed under hardware constraints and imperfect CSI, i.e., one direct-mapping algorithm suitable for high/moderate number of bits in PSs, one heuristic algorithm with improved receiving performance at the cost of complexity, and one iterative-closed-form algorithm with enhanced practicality of symbol-level based DM. Simulation demonstrates that the algorithms achieve lower symbol error rate (SER) at the LUs while significantly deteriorating the Eves’ SER, leading to an improved secrecy throughput over the benchmarks.

中文翻译：

---

使用少位移相器进行安全定向调制：优化和迭代闭合形式设计

在本文中，研究了定向调制 (DM) 以增强物理层安全性。实际发射机设计在不完善的信道状态信息 (CSI) 和硬件约束下进行开发，例如有限分辨率移相器 (PS) 和每个天线的功率预算。针对实现 DM 中的实际问题，研究了一系列实际场景。从窃听者（Eve）的信息完全未知的场景出发，提出相应的设计来优化合法用户（LU）的接收性能，同时随机化Eves的接收信号。当 Eves 的 CSI 不完全知道时，在第二种情况下，通过对 Eves 施加破坏性干扰，Eves 的接收性能进一步恶化。对于每个场景，在硬件约束和不完善的CSI下提出了三种算法，即一种适用于PS中高/中等比特数的直接映射算法，一种以复杂度为代价提高接收性能的启发式算法，以及一种迭代闭式算法增强了基于符号级的 DM 的实用性。仿真表明，这些算法在 LU 上实现了较低的符号错误率 (SER)，同时显着降低了 Eves 的 SER，从而提高了基准测试的保密吞吐量。