基于相位噪声加性高斯化的全双工极化自干扰消除研究

To break the bottleneck limit caused by phase noise, this proposal investigates polarization based phase noise Gaussianization for self-interference cancellation in full-duplex system. Taking advantage of polarization signal processing, the multiplicative phase noise will be transformed to an additive complex Gaussian noise. Hence, the self-interference signal can be cancelled effectively to noise floor without any prior knowledge of the phase noise. The research will be developed on the focus of polarization based phase noise Gaussianization and self-interference cancellation to optimize the data rate of full-duplex system as following. Firstly, the methods to control polarization of the full duplex system signals are investigated. Meanwhile, the principle of transforming phase noise to the additive Gaussian process is fully demonstrated; the full duplex polarization self-interference signal considering phase noise is built, and the cancellation schemes under different channel conditions are then researched. Secondly, the rate gain of the full duplex compared with the half duplex is researched, the upper bound of the gain is derived and corresponding scheme is proposed to optimize the rate gain. Rely on this project, the critical problem of limiting the self-interference reduction in traditional cancelation, which is due to phase noise, will be solved. A theory system, based on additive Gaussianized phase noise to reduce the polarized self-interference, is expected to be creatively established in the field of full duplex in wireless communication. This research will enable to double the spectrum efficiency further with approaching the theoretic upper bound limitation of self-interference cancellation.

为突破全双工中乘性相位噪声对自干扰消除性能的限制瓶颈，课题利用全双工信号所固有的极化属性，开创性地提出通过极化信号处理将乘性相位噪声进行加性高斯化处理来提升自干扰消除性能的思想。课题围绕如何实现加性高斯化相位噪声影响下的极化自干扰消除展开。首先研究全双工信号的极化状态控制机制并充分论证相位噪声加性高斯化特性，进而构建具有相位噪声的全双工极化自干扰信号模型，在此基础上研究不同信道条件下的极化自干扰消除方案，最后对极化全双工系统相比于半双工系统的数据速率增益进行研究，推导增益上限并提出优化增益性能的方案。课题从根本上解决了全双工传统自干扰消除受限于乘性相位噪声的科学难题，力争使自干扰消除性能逼近于理论上限，向实现全双工频谱效率倍增的目标迈出坚实的一步。

针对相位噪声影响全双工系统自干扰消除性能这一关键问题，项目围绕如何利用极化信号处理技术实现相位噪声加性高斯化展开，从全双工虚拟极化处理机制及相位噪声加性高斯化特性、对抗相位噪声的全双工极化自干扰消除算法、极化全双工系统数据速率增益等方面进行研究。研究成果为发表SCI检索论文4篇和 EI检索论文19篇，申请发明专利8项并获得1项授权，培养研究生20余人。. 课题首先从理论和仿真两个方面验证了基于极化信号处理方式可以将乘性相位噪声加性高斯化这一特性；在全双工极化自干扰信号建模基础上，研究了相位噪声和信道去极化效应对极化自干扰信号的影响；针对理想信道条件与实际去极化信道条件，分别提出了基于酉矩阵旋转与基于极化斜投影等一系列全双工自干扰消除算法；通过结合MIMO空间特性，实现了极化全双工系统数据速率的大幅提升；基于USRP与GNU Radio搭建了极化全双工通信平台，并验证所提算法性能。与传统时频域的自干扰消除算法性能对比，项目所提基于相位噪声加性高斯化的全双工极化自干扰消除算法在自干扰消除量上有5-10dB的性能增益，同时可以大幅提升数据速率。. 项目开创性地提出了基于相位噪声加性高斯化的全双工极化自干扰消除研究，将相位噪声的乘性影响转化为加性高斯噪声，为模拟振荡器相位噪声所导致的全双工自干扰消除受限问题开辟了新的数字化解决途径。项目研究创建了全双工极化自干扰消除新方法，给出了极化全双工系统数据速率增益的理论上限和实际信道中相应的优化方法，逐渐在无线通信领域建立了一个基于极化信号处理解决全双工自干扰消除问题的新理论体系。