时变信道分布式MIMO系统多信道参数估计和补偿技术

发布时间：2019-05-18 13:06

【摘要】：作为长期演进计划(Long Term Revolution,LTE)中的核心技术之一,多输入多输出(Multiple Input Multiple Output,MIMO)可以有效的提供分集增益和功率增益。随着对MIMO技术研究的深入,分布式MIMO系统以其高容量、广覆盖、低功耗、和易扩展的优势受到了广泛的关注和研究,成为未来移动通信的关键技术之一。在分布式MIMO系统中,由于收发天线可能分布在不同的地理位置,从而导致各收发天线对之间的频偏可能各不相同。高速铁路的飞速发展迫切需要适用于时变条件下的移动通信技术,因此研究分布式MIMO系统的多频偏估计问题并将这一问题扩展到时变条件下进行研究具有较为重要的意义。单输入单输出(Single Input Single Output,SISO)系统和集中式MIMO系统在接收端可以简单有效的补偿频偏,而分布式MIMO系统中发端天线存在多个不同频偏使得接收端难以对频偏进行最优补偿,因此有必要研究基于预编码技术的多频偏预纠正方法。本文第二章研究了准静态条件下分布式MIMO系统的多参数估计算法。首先,理论推导了分布式MIMO系统中多频偏估计的最大似然估计模型及估计的克拉美-罗限。然后,推导了基于训练序列相关的多频偏估计算法,针对此算法存在均方误差(Mean Square Error,MSE)平台的缺点,分析并推导了基于期望最大化(Expectation Maximization,EM)类的迭代多参数最大似然估计算法。最后,通过仿真分析了基于训练序列相关的估计算法和EM类算法的性能,并进一步对比了EM类算法中期望最大化(Expectation Conditional Maximization,ECM)算法和空间交替期望最大化(space-Alternating Generalized Expectation-maximization,SAGE)算法的性能。理论分析和仿真结果证明,基于EM类迭代的最大似然类算法在较高信噪比时可以较好地克服多天线间干扰,从而较好的克服了MSE平台问题,取得较好的多参数估计性能;相对于ECM算法,SAGE算法可以较快的达到收敛。本文第三章研究了时变条件下分布式MIMO系统的多参数估计算法。首先,研究了时变条件下分布式MIMO系统中多频偏估计的最大似然估计模型。其次,将ECM迭代算法推广到时变条件下的多频偏估计中,并完成理论推导。然后,针对ECM算法收敛速度慢的缺点,推导出时变信道条件下SAGE算法,该算法将噪声与隐藏数据空间关联以减少数据空间的费歇尔信息从而提高收敛速度,完成理论推导。理论分析和仿真结果证明,本章所提算法可以较好的克服信道的时变性,取得较好的多参数估计性能。本文第四章研究了时变条件下分布式MIMO系统的多频偏和天线增益的联合预校准技术。首先,研究了基于预编码的多频偏纠正技术。其次,在研究单天线校准技术的基础上研究了时变条件下的多天线校准模型及校准过程。然后,针对单独进行多频偏预纠正或天线校准都需要反馈信息的问题,把多频偏看做等效信道的一部分,使其与信道信息统一成为等效信道,并对此等效信道进行天线校准,从而可以在不增加反馈量的条件下同时实现多频偏预纠正和天线校准。理论推导和仿真结果证明,多频偏和天线增益的联合预校准技术可以在不增加反馈量的条件下同时实现多频偏预纠正和天线校准,取得较好的系统性能。综上,时变信道中多参数估计和补偿技术是分布式MIMO系统的关键技术之一,本文对分布式MIMO系统的多参数估计作了较深入的研究,并研究了联合多频偏和天线增益预校准的技术,具有一定的理论研究和实际应用价值。
[Abstract]:As one of the core technologies in Long Term Evolution (LTE), multiple-input multiple-output (MIMO) can effectively provide diversity gain and power gain. With the deep research of MIMO technology, the distributed MIMO system has received extensive attention and research with its high capacity, wide coverage, low power consumption and easy expansion, and becomes one of the key technologies for future mobile communication. In a distributed MIMO system, the frequency offset between the transmit and receive antenna pairs may vary due to the fact that the transmit and receive antennas may be distributed in different geographic locations. The rapid development of high-speed railway urgently needs to be applied to the mobile communication technology under time-varying conditions, so it is of great significance to study the multi-frequency offset estimation problem of the distributed MIMO system and to extend the problem to the time-varying conditions. The single-input single-output (SISO) system and the centralized MIMO system can simply and effectively compensate the frequency offset at the receiving end, Therefore, it is necessary to study the multi-frequency offset pre-correction method based on the precoding technique. In the second chapter, the multi-parameter estimation algorithm of the distributed MIMO system under quasi-static condition is studied. First, the maximum likelihood estimation model of multi-frequency offset estimation in a distributed MIMO system and the estimated KRA-Lo limit are derived. Then, a multi-parameter maximum likelihood estimation algorithm based on the expectation maximization (EM) class is derived. Finally, the performance of the estimation algorithm based on the training sequence and the EM class algorithm is analyzed by the simulation, and the performance of the expectation maximization (ECM) algorithm and the space-alternate generation (SAGE) algorithm in the EM algorithm is further compared. The theoretical analysis and simulation results show that the maximum likelihood algorithm based on the EM class iteration can overcome the inter-antenna interference better when the higher signal-to-noise ratio is higher, so that the problem of the MSE platform is better overcome, and the better multi-parameter estimation performance is obtained; and compared with the ECM algorithm, The SAGE algorithm can achieve convergence faster. In the third chapter, the multi-parameter estimation algorithm of the distributed MIMO system under time-varying conditions is studied. First, the maximum likelihood estimation model of multi-frequency offset estimation in a distributed MIMO system under time-varying conditions is studied. Secondly, the ECM iterative algorithm is extended to the multi-frequency offset estimation under time-varying conditions, and the theoretical derivation is completed. Then, for the disadvantage of slow convergence speed of the ECM, the SAGE algorithm under the condition of time-varying channel is derived. The algorithm associates the noise with the hidden data space to reduce the time-varying information of the data space, so as to improve the convergence speed and complete the theoretical deduction. The results of theoretical analysis and simulation show that the algorithm proposed in this chapter can better overcome the time-changing of the channel and achieve better multi-parameter estimation performance. In the fourth chapter, the combined pre-calibration technique for multi-frequency offset and antenna gain of a distributed MIMO system under time-varying conditions is studied. First, a multi-frequency offset correction technique based on pre-coding is studied. Secondly, the multi-antenna calibration model and the calibration process under time-varying conditions are studied on the basis of the single-antenna calibration technology. then, aiming at the problem that the feedback information is needed for the independent multi-frequency offset pre-correction or the antenna calibration, the multi-frequency offset is regarded as a part of the equivalent channel, and the multi-frequency offset is unified with the channel information into an equivalent channel, and the equivalent channel is subjected to antenna calibration, So that the multi-frequency offset pre-correction and the antenna calibration can be realized at the same time without increasing the feedback amount. The theory and simulation results show that the combined pre-calibration technique of the multi-frequency offset and the antenna gain can realize the multi-frequency offset pre-correction and the antenna calibration without increasing the feedback amount, so as to obtain better system performance. In this paper, the multi-parameter estimation and compensation technique in the time-varying channel is one of the key technologies of the distributed MIMO system. The multi-parameter estimation of the distributed MIMO system is studied in this paper, and the technology of the combination of the multi-frequency offset and the antenna gain pre-calibration is also studied. And has certain theoretical research and practical application value.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN919.3

【相似文献】