多波束卫星通信中基于压缩感知的预编码设计与干扰消除技术

Given the quick development of broadband satellite communication services, the frequency spectrum turns into a valuable resource now and in the future. Frequency reuse and frequency overlap can be helpful to the highly efficient usage of frequency resources. However, as a result Common Channel Interference (CCI) and Adjacent Channel Interference (ACI) become serious in satellite communication system, which causes the performance deterioration of satellite communication link. In this proposal, we will study the precoding techniques and iterative interference cancellation techniques based on compressive sensing (CS) theory in multibeam satellite communications. First, the channel estimation techniques based on the MMSE criterion will be researched when multiple signals arrive at the receiver synchronously or asynchronously. Due to the sparsity of satellite channel, simultaneous orthogonal matching pursuit (SOMP) and adaptive step size-sparsity adaptive matching pursuit (AS-SAMP) channel estimation methods are to be studied. Second, we will design the precoders under the power constraint condition and the hybrid baseband and RF precoding technique will be researched on account of the OMP channel estimation. Then, the interference cancellation technique is going to be researched based on the factor graph and the algorithms of interference cancellation using the MMSE criterion and the single-user matched filter are represented by adopting the factor graph. Finally, joint interference cancellation and synchronization method based on expectation maximization (EM) will be analyzed and designed. The multiple parameters estimation technique is going to be studied based upon the Maximum Likelihood criterion through Gauss-Newton iteration. Because of the sparse satellite channel, the multiuser detector will be studied based on joint spatial modulation and structured CS. The study of this proposal will contribute both the theoretic analysis and practical implementation of the efficient transmission in broadband satellite channels.

宽带卫星通信业务的迅速发展使得频谱成为宝贵的资源，频率复用和频谱叠加可以实现频谱的高效利用，但是增加了系统的共信道和邻信道干扰，导致通信系统性能下降。本课题研究多波束卫星通信中基于压缩感知的预编码技术与干扰消除技术。首先，研究信号同步/异步到达时的MMSE信道估计技术，针对稀疏卫星信道研究同时正交匹配追踪（SOMP）和自适应步进稀疏自适应匹配追踪（AS-SAMP）信道估计方法。其次，研究功率受限的预编码技术和采用OMP的信道估计方法的基带/射频混合预编码技术。然后，研究基于因子图的干扰消除技术，将最小均方误差准则和单用户匹配滤波干扰消除算法用因子图表示。最后，研究基于期望最大化的联合干扰消除和同步技术、基于Gauss-Newton迭代的最大似然多参数估计技术及基于联合空域调制和结构化压缩感知的多用户检测技术。本课题的研究对于宽带卫星信道的频谱高效传输具有重要的理论意义和实际应用价值。

宽带卫星通信业务的迅猛发展使得频谱成为宝贵的资源，频率复用和频谱叠加可以提高频谱的利用效率，但是增加了系统的共信道和邻信道干扰，导致通信系统性能下降。课题的主要研究内容包括：1）研究受雨衰影响的多波束卫星通信信道的建模和表示方法；2）研究抗干扰的多波束卫星通信的线性和非线性预编码技术；3）研究非线性信道条件下的卫星通信迭代检测技术。. 针对上述研究内容开展了研究，建立Ka波段受雨衰影响的多波束卫星通信信道模型，提出一种基于单天线功率约束和业务速率需求约束的线性预编码设计方案，将加权最大最小公平性函数作为优化目标，通过恰当地选取权重值，采用半定松弛方法和二分法联合求解优化目标的反问题得到最优解，在满足加权最大最小公平性条件下，该线性预编码方案的总吞吐量比现有速率平衡方案有显著提升；研究了一种满足服务质量（QoS）约束条件的MIMO下行链路非线性预编码方案，该算法基于最小均方误差（MMSE）准则，通过Cholesky分解得到预编码矩阵的闭式解，且对信道矩阵进行适当的排序，该非线性预编码方案的性能和可达速率比迫零（ZF）的非线性预编码方案有明显改善，发射机能够根据每个用户的QoS要求来发送不同类型的数据业务；研究了非线性信道中采用FTN传输的迭代接收机设计方法，利用自回归过程表示色噪声，将非线性信道建模为Volterra模型并进行线性化处理，基于因子图和消息传递框架设计低复杂度的检测方法，相比奈奎斯特成型，在相同复杂度下该FTN方案可以提升25%的传输速率。. 上述研究工作较好地解决了该项目拟解决的三个问题，完成了项目的研究任务，提出的预编码设计方法和频谱高效的迭代检测方法对提升下一代高吞吐量的卫星通信的性能具有一定的促进作用。