Robust Adaptive Coded Modulation for Time-Varying Channels

最近的工作表明，自适应信令具有极大地提高在无线信道上操作的数字通信系统的带宽效率的理论承诺。 然而，由于无线信道随时间的变化，自适应信令的效用受到质疑，这导致数据传输时的信道与信道估计时的信道不同。 PI最近的研究结果表明，这些问题不能忽视;信道变化极大地改变了问题的性质，因此，无线信道的鲁棒自适应信令方案的设计提出了显着不同的挑战，比有线信道的类似设计问题。 这些挑战中的许多是由这些算法中所需的鲁棒性驱动的，这是由于在发射机处关于信道衰落过程的统计的不确定性。 在这个项目中，进行了一个全面的研究时变信道的鲁棒自适应信令方案的设计。 本研究的主要部分包括：（1）采用给定的自适应信令范例的鲁棒算法的设计和表征;（2）非常适合于时变信道的自适应编码范例的开发;以及（3）自适应混合ARQ方案的开发，该方案具有通过重传部分保证鲁棒性的额外灵活性。 这项工作的耦合与必要的统计模型的发展，特别是适当的类的自相关函数的信道衰落过程的发展，将导致家庭的自适应编码策略，可以采用在系统中操作的时变信道与各种速率的信道变化。这些结果将说明自适应编码是有用的应用程序和它应该被采用的方式。 将用于获得这些结果的方法来自概率和随机过程，通信系统理论和统计信号处理领域。 特别是，结果在很大程度上依赖于随机过程理论的结果，编码调制方案的设计，和扩展的结果从最小最大鲁棒时间序列预测。

Recent work has suggested that adaptive signaling has the theoretical promise of greatly increasing the bandwidth efficiency of digital communication systems operating over wireless channels. However, the utility of adaptive signaling has been questioned due to the variation of the wireless channel over time, which results in a different channel at the time of data transmission than at the time of channel estimation. Recent results by the PI have shown that these issues cannot be disregarded; the channel variation greatly alters the nature of the problem and thus the design of robust adaptive signaling schemes for wireless channels presents significantly different challenges than the analogous design problem for wireline channels. Many of these challenges are driven by the robustness necessary in these algorithms due to uncertainties at the transmitter about the statistics of the channel fading process. In this project, a comprehensive study of the design of robust adaptive signaling schemes for time-varying channels is undertaken. The main portions of this study include: (1) The design and characterization of robust algorithms that employ a given adaptive signaling paradigm; (2) The development of adaptive coding paradigms that are well suited to time-varying channels; and (3) The development of adaptive hybrid-ARQ schemes, which have the added flexibility of partially guaranteeing robustness through retransmission. The coupling of this work with the development of the necessary statistical models, in particular the development of appropriate classes of autocorrelation functions for the channel fading processes, will lead to families of adaptive coding strategies that can be employed in systems operating over time-varying channels with various rates of channel variation. These results will illustrate both the applications where adaptive coding is useful and the manner in which it should be employed. The methods that will be used to obtain these results are drawn from the areas of pr obability and random processes, communication system theory, and statistical signal processing. In particular, the results rely heavily on results from random process theory, the design of coded modulation schemes, and extensions of results from minimax-robust time series prediction.