Turbo码是逼近香农极限的好码，以其独特的优势在无线和移动通信领域得到广泛的应用，成为第三代移动通信及其长期演进计划和第四代移动通信系统最重要的信道编码方式。自Turbo码诞生以来，其各种译码算法和实现架构均分量译码器贯序译码然后进行迭代的思路；该方法极大地制约了Turbo译码器速度的提升。本课题研究将以Turbo码外信息交换方法为切入点，以全并行Turbo译码算法架构、全并行译码算法的分析方法、全并行译码算法的基本性能分析等为核心研究问题，并讨论其基于概率计算的高效VLSI实现方法为研究重点。通过对这问题的研究，本课题将提出全并行的Turbo译码算法及其实现方法，并证明其有效性。本课题从信息论的角度出发，在继续保持Turbo码译码性能的基础上，突破传统译码算法及其实现方法对译码器吞吐量的制约，从而为未来高速、宽带无线/移动通信系统提供高吞吐量、高效的Turbo码译码算法及高效实现方法。

Turbo code is the good code for approaching the Shannon limitation, and has been widely used in wireless and mobile communications with its unique advantages. Turbo code is the most important channel coding scheme for the third generation mobile communication and its long time evolution (LTE) and fourth-generation mobile communication systems. The sequential decoding of component decoder and iteration processing between two component decoders are used in the decoding algorithms and implementation architectures of Turbo code since its birth. This processing method has become the bottleneck for the decoding speed of the turbo decoder. In this project, we propose a novel architecture for turbo decoding algorithm based on the parallel extrinsic information exchange to achieve fully parallel Turbo decoding. The full parallel Turbo decoding algorithm architecture, performance analysis and high efficient implementation scheme are all studied in this project. We demonstrate that the proposed algorithm architecture can achieve the same properties and performance of the traditional Turbo decoding methods with high throughput. This topic will break through the bottleneck of the traditional decoding algorithm and its implementation form the view point of methodological. Thus, we will provide support for the future of high-speed, broadband wireless/mobile communications system with high throughput and high efficient turbo decoding algorithm and efficient implementation.