Secrecy Rate Optimizations for a MIMO Secrecy Channel With a Multiple-Antenna Eavesdropper
Secrecy Rate Optimizations for a MIMO Secrecy Channel With a Multiple-Antenna Eavesdropper
复制标题
DOI：
10.1109/tvt.2013.2285244
复制
发表时间：
2014-05-01
影响因子：
6.8
通讯作者：
Leung, Kin K.
中科院分区：
文献类型：
--
作者：
Cumanan, Kanapathippillai;Ding, Zhiguo;Leung, Kin K.
研究了多输入多输出（MIMO）保密信道中不同的保密率优化问题。特别是，我们考虑的情况下，通过MIMO信道的通信被偷听的多天线窃听。在这个保密网络中，我们首先研究两个保密率优化问题：1）功率最小化和2）保密率最大化。这些优化问题是不凸的，由于非凸的保密率约束。然而，通过近似基于泰勒级数展开的保密率约束，我们提出了迭代算法来解决这些保密率优化问题。此外，我们提供的算法的收敛性分析。这些迭代优化方法是在假设发射机具有完美的信道状态信息的情况下开发的。然而，在发射机处具有完美的信道状态信息存在实际困难。因此，鲁棒的保密率优化技术的基础上，最坏情况下的保密率被认为是通过将信道的不确定性。通过利用S-过程，我们表明，这些强大的优化问题可以制定成半定规划在低信噪比（SNR）。仿真结果验证了算法的收敛性。此外，数值结果表明，所提出的鲁棒优化技术优于非鲁棒计划的最坏情况下的保密率和实现的保密率。
This paper studies different secrecy rate optimization problems for a multiple-input-multiple-output (MIMO) secrecy channel. In particular, we consider a scenario where a communication through a MIMO channel is overheard by a multiple-antenna eavesdropper. In this secrecy network, we first investigate two secrecy rate optimization problems: 1) power minimization and 2) secrecy rate maximization. These optimization problems are not convex due to the nonconvex secrecy rate constraint. However, by approximating this secrecy rate constraint based on Taylor series expansion, we propose iterative algorithms to solve these secrecy rate optimization problems. In addition, we provide the convergence analysis for the proposed algorithms. These iterative optimization approaches are developed under the assumption that the transmitter has perfect channel state information. However, there are practical difficulties in having perfect channel state information at the transmitter. Hence, robust secrecy rate optimization techniques based on the worst-case secrecy rate are considered by incorporating channel uncertainties. By exploiting the S-Procedure, we show that these robust optimization problems can be formulated into semidefinite programming at low signal-to-noise ratios (SNRs). Simulation results have been provided to validate the convergence of the proposed algorithms. In addition, numerical results show that the proposed robust optimization techniques outperform the nonrobust schemes in terms of the worst-case secrecy rates and the achieved secrecy rates.