【作者】 李丹丹；

【导师】 裘正定；

【作者基本信息】 北京交通大学 ， 信号与信息处理， 2014， 博士

【摘要】 协作通信系统通过多个单天线节点共享彼此的天线以形成一个虚拟的多天线系统。在双向多中继系统中应用协作通信技术可以提高系统的容量、安全性、鲁棒性,降低系统的误码率和中断概率,已经成为了当前的一个研究热点。放大转发(AF)和解码转发(DF)是中继对信息的两种主要处理方式。本文首先分析了DF双向单中继系统的误符号率(SER)性能,并以此为基础研究了DF双向多中继系统的中继选择方案。接着分别以提高系统和速率与提高系统安全性为目标,设计了AF双向多中继系统的中继波束成形向量。本文的主要研究内容为：首先,扩展了传统最近邻近似方案,且应用该方法对DF双向单中继系统的中继端SER进行讨论,得到了加性高斯白噪声(AWGN)信道下BPSK调制、QPSK调制、8PSK调制的SER闭式表达式和衰落信道下BPSK调制、QPSK调制的瞬时SER与平均SER的表达式。研究发现,在AWGN信道下采用8PSK调制的DF双向中继系统中,固定其中一个源节点的发送功率,另一个源节点的功率从0增大到与其功率相等的过程中,中继节点的SER并不是单调递减的。其次,在DF双向单中继系统SER性能分析的基础上,将场景扩展到双向多中继系统,提出了一种基于最小SER的DF双向多中继系统中继选择方案。考虑到最小SER中继选择方案的复杂度较高以及中继星座图中最近邻星座点之间的距离在端到端SER中的重要作用,进一步提出了基于星座图最小距离最大化的中继选择方案,并推导出在BPSK、QPSK和8PSK调制时,该方案能达到满分集增益。然后,研究了AF双向中继系统的信息传输和能量收集问题。即在中继功率约束以及源节点能量收集约束下,通过设计中继的波束成形向量从而最大化系统的和速率。本文分别提出了一种理想情况设计方案、三种实际设计方案和一种鲁棒设计方案,并对这些方案建立了最优化模型,且通过半正定松弛以及连续凸近似技术对它们进行求解,还证明了所提出的求解算法至少可收敛到一个局部最优解。最后,研究了有窃听节点存在时AF双向中继系统物理层安全性问题。提出了一种中继波束成形设计方案以及一种中继波束成形与人工噪声的联合设计方案,从而在保证双向中继系统的通信质量和安全性的前提下最小化中继的功率和,并为这两个方案建立了最优化模型,且应用半正定松弛技术对这两个非凸的优化问题进行了求解。证明了在所提出的两种方案中,应用人工噪声方案的性能更优。更多还原

【Abstract】 In a cooperative communication system, a virtual multiple-antennas environment is constructed by sharing antennas of different nodes, and the diversity gain can be achieved without increasing the hardware complexity. Applying the cooperative techniques to two-way multi-relay system can provide high capacity, low symbol error rate and low outage probability. Amplify-and-forward (AF) and decode-and-forward (DF) are two main re-laying protocols in a two-way multi-relay system. In this paper, we investigate the S-ER performance of a two-way relay system firstly, and then propose two relay selection schemes for the DF based two-way relay system. Moreover, in order to improve the sum-rate and the security of an AF based two-way relay system, the relay beamforming design is explored.Firstly, in order to discuss the SER of a DF based two-way relay system, we extend the nearest neighbor approximation method. By this method, the SER performance with additive white Gaussian noise channel for BPSK, QPSK and8PSK modulation is derived. For fading channel, the constellation at relay is complex due to scaling and rotation of channel fading. We discuss the instantaneous SER at first, and then by calculating the distribution of the shortest distance between the constellation points, the average SER for BPSK and QPSK modulation over Rayleigh fading channels is derived.Secondly, a relay selection scheme is proposed to improve the SER performance of a two-way relay system. Since the shortest distance between points in the constellation is an important factor for the end-to-end SER of a two-way relay system, we propose another relay selection scheme, in which the relay with the maximal minimum distance between points in the constellation will be selected. When BPSK, QPSK and8PSK modulation are applied, full diversity gain can be achieved. Moreover, a look-up table based method is also designed for high order modulation.Thirdly, the sum-rate maximization with transmit power and energy harvesting con-straints are considered in the two-way AF based multi-relay system. We design an ideal scheme, three practical schemes and one robust scheme. The optimization problems are solved by semidefinite relaxation and successive convex approximation method. We also prove that the proposed algorithms at least achieve the local optimal solution.Finally, the problem of secure information exchange between two sources via mul- tiple relays in the presence of an eavesdropper is discussed. To this end, we consider both cases with and without artificial noise and formulate two optimization problems for beamforming design. The goal is to seek the optimal beamforming vectors to minimize the total power consumed by relay nodes such that the secrecy of the information ex-change between the two sources can be protected. Since both optimization problems are nonconvex, we solve them by semidefinite relaxation method, and prove that the beam-forming with artificial noise scheme outperforms the relay beamforming without artificial noise in terms of both low power consumption and low infeasibility rate.更多还原