Turbo码瀑布区性能建模和性能估计是构建完整的Turbo码性能分析框架的重要一环，能够同时刻画和描述Turbo码瀑布区和错误平层两个特征性能区域的完整Turbo码性能建模也是Turbo码优化设计、乃至评估采用Turbo码通信系统性能的重要基石，鉴于目前缺乏有效的Turbo码瀑布区性能分析方法，围绕瀑布区Turbo码迭代译码定量建模与性能分析的相关工作具有重要的理论和应用价值。本项目针对瀑布区Turbo码性能建模和性能估计所面临的关键问题，重点研究和揭示Turbo码瀑布区性能形成机理，并从迭代译码输出对数似然比LLR概率分布与统计特性分析入手，研究推导刻画Turbo码瀑布区的性能模型以及相关模型参数的定量计算方法。此外，项目还将结合能够同时刻画瀑布区和错误平层区的Turbo码性能模型，研究优化的Turbo编译码方法。项目研究成果可以推广到一般性地分析复杂迭代译码/检测机制性能。

The waterfall region performance modeling and performance estimate plays an important role in the performance characterization for Turbo codes. And the complete performance estimate of both the waterfall region and the error floor region provides the basis for the optimized Turbo code design, and even the foundation to enable comprehensive system performance assessment for communication system with Turbo codes. Since there is few knowledge about the accurate waterfall region performance model for Turbo codes owing to the complicated iterative decoding scheme utilized, the modeling as well as the quantitative analysis of the waterfall region performance for Turbo codes not only is of important theorectical significance, but also have important engineering value. This research project aims at the critical problems in waterfall region performance modeling and performance estimate for Turbo codes. We will focus on the formulation mechanism analysis of the waterfall region performance characteristics for Turbo codes. Moreover, we will try to derive the performance model and find out how to determine the involved parameters to characterize the waterfall region of Turbo codes by analyzing the statistical property and stochastic probability distribution of the log-likelihood ratio (LLR) in iterative Turbo decoding. In addition, we will investigate the optimized Turbo code design based on both the waterfall region and the error floor region performance characterization for Turbo codes. Finally, the research results may hopefully shed lights on the generalized performance estimate for other complicated iterative decoding and iterative detection mechanism.