低信噪比条件下数字通信系统码辅助同步技术研究

High data rate transmission technologh in DRSS becomes an active reaearch field in wireless communication. With the rapid increase of transmission capacity, high data rate transmission technologh is faced with the double pressure of limited spectrum and power resource, which can be released by combining the high order modulation schemes with efficient channel codes. The application of efficient channel codes such as Turbo and LDPC codes can make the decoding threshold falling and power efficiency increasing. However, it also poses a great challenge to signal demodulation especially for traditional synchronization at low signal-to-noise ratios (SNR) to make the powerful coding gain come true. With the view of improving the estimation performance of synchronization parameters at low SNR in this research subject, the framework for joint iterative code-aided synchronization is managed to construct, which involved several aspects, such as symbol timing, carrier synchronization and frame synchronization. Firstly, the coarse estimation of synchronization parameters will be solved by comparing the presetted threshold with a cost function which is constructed by the soft information from iterative decoder. Secondly, the soft information will be introduced into a parameter estimation framework on the basis of factor graph and expectation-maximization (EM) algorithm to improve the estimation precision of synchronization parameters. Finally, the joint synchronization scheme with low computational complexity will be discussed based on the interdependence of different synchronization parameters. The findings may offer new guidance and theory foundation for the design of efficient synchronization system at low SNR.

中继卫星高速数据传输技术是无线通信领域的研究热点，也是发展我国空间信息网络必须解决的关键技术。随着传输速率的快速增长，高速数传面临着频谱/功率受限的双重压力，而采用高阶调制与高效信道编码技术相结合将是缓解这一压力的有效途径。虽然Turbo/LDPC码的应用能极大降低译码门限，提高功率利用率，却给信号解调尤其是传统同步过程带来了巨大挑战，如何在低信噪比条件下完成有效同步成为实现高效编码增益的关键因素。本项目以提高低信噪比同步参数估计性能为切入点，试图从符合定时、载波同步和帧同步等方面建立码辅助的联合迭代同步框架。首先利用迭代译码过程产生的软信息构建代价函数，通过门限检测方法完成同步参数粗估计；然后将软信息引入到因子图和EM参数估计框架，改善同步参数估计精度；最后，根据不同参数之间的相互依赖关系，探讨一种低复杂度联合同步方案。研究成果将为低信噪比条件下高效同步系统的设计提供新思路和理论依据。

为解决低信噪比条件下数字通信系统传统同步方法难以高效工作的问题，本课题主要研究码辅助符号定时、载波同步、帧同步以及联合同步等内容，主要研究结果如下：.（1）为有效应对符号定时偏差较大时，现有码辅助定时算法出现有偏估计的问题，本课题提出了基于双路并行M值搜索和EM精估计级联的码辅助符号定时算法。该算法不仅极大地降低了实现复杂度，并且具有理想的RMSE性能和BER性能。.（2）针对M值搜索载波同步方法复杂度太高，而EM算法估计范围太小的问题，本课题提出了基于M值搜索粗估计和EM精估计的码辅助载波同步算法。为降低算法复杂度，我们利用载波相偏以及载波频偏的改变与M值的变化存在某些特殊的规律来完成粗估计，并对CA-Fitz频偏估计算法和CA-V&V相偏估计算法进行了简化设计。仿真结果表明：所提的载波同步方案的RMSE性能和BER性能相对于理想载波同步情况下的性能损失很小。.（3）本课题提出了一种码辅助帧同步算法，该算法相比于传统相关峰检测和EM帧同步算法而言，缩短了对帧同步码长度，提高了系统频谱效率；相比于M值搜索帧同步算法而言，则极大地降低了实现复杂度。.（4）本课题还提出了一种码辅助联合同步方案，该方案由M值搜索粗同步和EM精同步算法构成。为了加快EM算法收敛速度，我们在EM符号定时之前进行载波频偏和相偏预先校正。仿真表明该联合同步方案的系统BER性能相对于理想同步时的性能损失小于0.5 dB。.研究结果表明，在低信噪比条件下，通过合理利用LDPC译码器输出软信息，可以显著提升符号定时、载波同步以及帧同步的性能，进而改善系统BER，为低信噪比条件下LDPC编码系统的高效应用提供了参考方案。