Analysis and design of iterative (turbo) digital receivers

本项目的目标是开发用于分析数字接收机中采用的迭代（所谓的“turbo”）处理的方法，作为产生接近最佳性能的手段，即在不利条件下几乎无差错的通信。这一建议的及时性是由于数字通信深刻融入现代生活的重要性日益增加。新的技术解决方案是生产具有成本效益的产品的“必须”，因此研究工作不断部署，以设计数字收发器。这项研究与所谓的turbo码的发明有关，turbo码的迭代处理的突破性思想迅速扩展到许多应用中。使用“turbo”原理，接收器被划分为相对小的块（设备），在本地处理数据，并且迭代地交换可靠性信号以接近全局最优处理。对turbo处理越来越感兴趣的原因是，它可以产生接近最佳的解决方案，并且具有可管理的复杂性，使实际接收机的性能接近其理论极限。因此，分析迭代算法的理论和实际问题并研究其应用是非常重要的。 目前，各种“涡轮”算法被提出，但正式分析它们的工具是稀缺的。这是因为迭代处理是高度非线性的，并且在大多数情况下抵抗形式分析，这极大地限制了研究人员设计新接收器的能力。这为我们的研究提供了动力。我们的目标是设计分析或半分析工具，以获得接收器的组成设备之间交换的消息的概率模型，具有简单和最佳设计等优点，以及接收器对时变信道条件的适应性。资金将用于支持研究生和传播研究成果。

The objective of this project is to develop methods for analysis of iterative (the so-called "turbo") processing employed in digital receivers as a mean to yield close-to-optimal performance, i.e. almost-error-free communication in adverse conditions. The timeliness of this proposal is due to ever-growing importance of digital communications integrating profoundly the modern life. New technological solutions are "a must" to yield cost-effective products, so research efforts are continuously deployed to design digital transceivers. This research is related to the invention of the so-called turbo-codes, whose break-through idea of iterative processing was quickly extended on many applications. Using the "turbo" principle, the receiver is divided into relatively small blocks (devices) processing the data locally, and exchanging the reliability signals iteratively to approach the globally optimal processing. The reason behind the growing interest in turbo processing is that it yields close-to-optimal solutions with manageable complexity, bringing the performance of practical receivers close to its theoretical limits. It is, therefore, important to analyze theoretical and practical aspects of the iterative algorithms and study their applications.  Currently, various "turbo" algorithms are being proposed but the tools to analyze them formally are scarce. This is because the iterative processing is highly non-linear and resists formal analysis in a majority of cases which significantly limits the researchers' capacity to devise new receivers. This provides motivation for our research. Our objective is to devise analytical or semi-analytical tools to obtain probabilistic models of messages exchanged between the constituent devices of the receiver, with benefits such as simple and optimal design, as well as adaptation of the receivers to the time-varying channel conditions.The funding will be used to support graduate students and to disseminate the research results.