ITR: Adaptive Signaling and MIMO Precoding for Rapidly Time-Varying Fading Channels

本课题研究了宽带移动无线电通信系统中快速时变衰落信道发射机优化的可行性。自适应发射机优化方法有潜力通过调整传输的信号-衰落条件来显著改善无线信道中的性能。然而，快速的信道变化使得这些闭环技术在无线系统中的应用变得困难。以往的研究表明，当车速较大时，仅利用过时的信道状态信息（CSI）在自适应传输算法中是不够的。要在实际中实现这些方法，必须提前几十到几百个符号预测信道系数。一种新的自适应，远程衰落信道预测算法（LRP）先前由PI和她的合作者为窄带无线信道开发。在本研究中，将天线阵列的远程衰落预测能力扩展到频率选择性衰落信道。这些方法被用于多载波和多天线系统的自适应调制和编码技术，以及用于多输入多输出（MIMO）宽带移动无线电系统的新型预编码方法和联合发射机/接收机优化技术。该研究有助于自适应传输和预编码方法的开发和实现，这是可靠的高速率无线通信所必需的。此外，与工业界的合作有利于参与项目的研究生。由于PI和研究生都是女性，因此代表性不足的群体的更广泛参与得到了加强。通过将本研究中调查的方法纳入高年级和班级项目，并与北卡罗来纳州立大学的无线通信活动相结合，丰富了研究和教育的基础设施。该项目的第一个组成部分侧重于利用预测方法消除多载波、多天线和多用户直接顺序码分多址（DS/CDMA）信道下行链路中出现的MIMO信道的干扰。虽然干扰源取决于应用，但这些系统可以用单个底层MIMO模型来描述。针对该模型，提出了新的tomison - harashima预编码方法。这些方法有望显著改进先前提出的线性预编码技术。此外，它们根据应用程序的要求在发送器和接收器之间分配计算资源。对于DS/CDMA信道的下行链路，研究了一种新的低复杂度线性技术，这种技术有可能大大简化基站的预编码和移动站的检测。所提出的干扰消除技术与自适应调制和编码相结合。由于这些预编码和自适应传输方法依赖于发射机中准确的CSI，因此可靠的LRP和潜在的预测误差统计在所提出方法的设计和性能分析中至关重要。在第二部分中，研究了多载波系统中多子载波和多天线信道中天线阵列单元的联合预测。探索了新的线性和非线性方法进行长期预测。建立了预测误差的统计模型，并将其用于设计可靠的自适应调制和编码方法。探索了以频谱效率换取复杂性和反馈要求的灵活预测方法。

This project investigates feasibility of transmitter optimization for rapidly time varying fading channels encountered in wideband mobile radio communication systems. Adaptive transmitter optimization methods have the potential to significantly improve performance in wireless channels by adjusting the transmitted signal-to-fading conditions. However, rapid channel variation makes application of these closed-loop techniques difficult in wireless systems. Previous investigations demonstrate that it is not sufficient to utilize the outdated Channel State Information (CSI) in adaptive transmission algorithms when vehicle speeds are significant. Prediction of the channel coefficients several tens-to-hundreds of symbols ahead is essential to realize these methods in practice. A novel adaptive, long-range fading channel prediction algorithm (LRP) was previously developed by the PI and her collaborators for narrowband wireless channels. In this research, the long range fading prediction capability is extended to frequency selective fading channels with antenna arrays. These methods are utilized in adaptive modulation and coding techniques for multicarrier and multiple antenna systems and in novel precoding methods and joint transmitter/receiver optimization techniques for Multiple Input Multiple Output (MIMO) wideband mobile radio systems. This research contributes to the development and realization of adaptive transmission and precoding methods that are essential in reliable high rate wireless communication. Moreover, collaboration with industry benefits graduate students involved in the project. Broader participation of underrepresented groups is enhanced since the PI and the graduate students are females. The infrastructure for research and education is enriched by incorporating the methods investigated in this research in senior and class projects and by integrating with wireless communications activities at NC State. The first component of this project focuses on utilization of prediction methods in interference cancellation for the MIMO channels that arise in multicarrier, multiple antennae and the downlink of multiuser Direct Sequence Code Division Multiple Access (DS/CDMA) channels. While the sources of interference depend on the application, these systems can be described with a single underlying MIMO model. Novel Tomlison-Harashima precoding methods are developed for this model. These methods are expected to improve significantly on previously proposed linear precoding techniques. Moreover, they distribute computational resources between the transmitter and the receiver as dictated by the application. For the downlink of the DS/CDMA channel, novel low complexity linear techniques that have the potential to significantly simplify precoding at the base station and detection at the mobile station are investigated. The proposed interference cancellation techniques are combined with adaptive modulation and coding. Since these precoding and adaptive transmission methods depend on accurate CSI in the transmitter, reliable LRP and the underlying prediction error statistics are essential in the design and the performance analysis of the proposed methods. In the second component, joint prediction for multiple subcarriers in multicarrier systems and antenna array elements in multiple antenna channels is investigated. Novel linear and nonlinear approaches to long-range prediction are explored. A statistical model of the prediction error is developed and utilized in the design of reliable adaptive modulation and coding methods. Flexible prediction methods that trade spectral efficiency for complexity and feedback requirements are explored.