In this paper, the secret-key capacity is maximized by optimizing the downlink training sequence in a time division duplexing (TDD) massive multiple-input-multiple-output (MIMO) scenario. Both single-user and multiple user cases are considered. As opposed to previous works, the optimal training sequence and the related secret-key capacity is characterized in closed-form in the single-user case and the large antenna multiple-user case. Designs taking into account a constraint on the maximal number of pilots are also proposed. In the multiple-user case, both the max-min and the sum capacity criteria are considered, including potential user priorities. In the end, it is shown that massive MIMO boosts the secret-key capacity by leveraging: i) spatial dimensionality gain and ii) array gain. Moreover, in the large antenna case, the multiple-user capacity is obtained with no extra pilot overhead as compared to the single-user case.
翻译:在本文中,通过在时间分解(TDD)大规模多投入-多产出(MIMO)情况下优化下链培训序列,使秘密钥匙能力最大化。考虑的是单一用户和多个用户的情况。与以往的工程相比,最佳培训序列和相关的秘密钥匙能力在单一用户案例中以封闭形式和大型天线多用户案例中具有特征。还提出了在设计中考虑到对试点项目最大数量的限制。在多用户案例中,考虑到最大数量和总能力标准,包括潜在的用户优先事项。最后,显示大型IMO通过杠杆作用(i) 空间多维度增益和(ii) 阵列增益来增强秘密钥匙能力。此外,在大型天线案例中,多用户能力是在与单一用户案例相比没有额外试点管理的情况下获得的。