LTE同频同时全双工自干扰抵消技术的研究

发布时间：2018-06-21 00:51

本文选题：同频同时全双工 + 天线抵消　；参考：《哈尔滨工业大学》2014年硕士论文

【摘要】：在无线通信领域,频谱是极其宝贵的资源,随着无线通信技术的发展,无线接入的总量成指数型增长,频谱资源越来越捉襟见肘。在一定频谱资源内,一个节点内能够接入更多的用户和能够满足每个用户更高的连接速度是我们追求的目标。而传统的双工方式和现行的频分双工、时分双工,都是以牺牲效率或者频谱为代价的,而不能完成真正意义的全双工。在学术界研究者们长期认为:在同一频带内,一个节点不能同时完成发送信号和接收信号,因为此节点的发射信号会对接收端造成很强的同频自干扰。而从近两三年的研究发现,同频同时全双工(Co-frequency Co-time Full Duplex,CCFD)是很有可能实现的。因此,本文研究了同频同时全双工的通信方式,它能实现在同一频带内,一个节点能够同时进行发送信号和接收信号。并采用了天线干扰抵消、模拟干扰抵消和数字干扰抵消等多层次联合干扰消除的体系架构,来完成对本地同频自干扰的抵消,从而实现对频谱资源的更高效的利用。本文首先研究了CCFD技术对现有技术的优势及其实现的关键问题和难点:抵消本地同频自干扰,并研究了本地同频自干扰经过收发机元件和无线信道的成分变化。本文研究了三天线实现天线抵消的技术,同时引入两路信号幅度不匹配和天线RX放置误差来表示工程中的误差对三天线抵消性能的影响,并通过仿真给出其误差对性能影响。为了进一步降低本地自干扰信号功率,本文还研究了模拟抵消的技术。采用一种类似于sinc差值的模拟抵消方法,利用本地模拟信号,构建N路等时延间隔的信号,在接收端通过控制接收信号功率最小来调整N路信号的衰减系数,将N路信号合成与自干扰抵消。为了充分抵消本地自干扰信号,本文最后研究了非线性数字干扰抵消算法。首先,对本地自干扰线性成分进行分析并建模,对其非线性部分分析并建模为泰勒级数,发现其偶次高阶项可以忽略。通过把PA功率放大器建模为并联汉默斯坦模型,并对天线抵消、数字抵消建模成FIR滤波器,完成基带非线性自干扰信道建模,得到自干扰信号。利用本地数字基带信号构造非线性变换矩阵,在数字接收端控制接收信号功率最小,利用最小二乘估计出变换矩阵系数,将构造的信号与自干扰合成完成抵消。最后对上述方法进行了实验仿真,并分析了仿真结果。
[Abstract]:In the field of wireless communication, the spectrum is an extremely valuable resource. With the development of wireless communication technology, the total amount of wireless access has increased exponentially, and the spectrum resources are increasingly stretched. In a certain spectrum resource, it is our goal to be able to access more users within a node and to meet the higher connection speed of each user. The traditional duplex mode and the current frequency division duplex and time division duplex are all at the expense of efficiency or spectrum but can not complete the true meaning of full duplex. Researchers in academic circles have long believed that in the same frequency band, a node can not complete the transmission signal and receive signal simultaneously, because the transmitted signal of this node will cause strong co-frequency self-interference to the receiver. In the past two or three years, it has been found that co-frequency Co-time full Duplex CCFDs are very possible. Therefore, this paper studies the communication mode of simultaneous full duplex in the same frequency band. It can be realized in the same frequency band, and a node can transmit and receive signals at the same time. The multi-level joint interference cancellation architecture such as antenna interference cancellation, analog interference cancellation and digital interference cancellation is adopted to cancel the local co-frequency self-interference, so as to achieve more efficient use of spectrum resources. This paper first studies the advantages of CCFD technology to the existing technology and the key problems and difficulties in its implementation: canceling local co-frequency self-interference, and studying the component changes of local co-frequency self-interference through transceiver components and wireless channels. In this paper, the technique of antenna canceling by three antennas is studied. At the same time, two signal amplitude mismatches and antenna RX placement errors are introduced to express the effect of the errors on the performance of the three antennas, and the effects of the errors on the performance are given by simulation. In order to further reduce the local self-interference signal power, this paper also studies the technology of analog cancellation. An analog cancellation method similar to sinc difference is used to construct N-channel signal with equal delay interval by using local analog signal. The attenuation coefficient of N-channel signal is adjusted by controlling the minimum power of received signal at the receiving end. N-path signal synthesis and self-interference cancellation. In order to fully cancel the local self-interference signal, the nonlinear digital interference cancellation algorithm is studied in this paper. Firstly, the local self-disturbance linear component is analyzed and modeled, and the nonlinear part is modeled as Taylor series. It is found that the even order higher order term can be neglected. The PA power amplifier is modeled as a parallel Hammerstein model, and the antenna cancellations and digital cancellations are modeled as Fir filters to model the baseband nonlinear self-interference channels and obtain the self-interference signals. The local digital baseband signal is used to construct the nonlinear transformation matrix, the power of the received signal is minimized at the digital receiver, the coefficient of the transformation matrix is estimated by the least square method, and the constructed signal is offset by the self-interference synthesis. Finally, the simulation results of the above methods are analyzed.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5

【参考文献】