快时变信道下OFDM系统频率同步与信道估计技术研究

发布时间：2019-04-04 13:00

【摘要】：随着用户数量的不断扩大以及多媒体业务的大量涌现,第三代移动通信系统已不能满足未来移动通信的发展需要,因此3GPP组织启动了3G技术的长期演进(Long Term Evolution, LTE)工作。OFDM作为LTE物理层的关键技术之一,不仅能够提供较高的数据传输速率,而且还具有较强的抗多径衰落和抗符号间干扰能力。但是,OFDM在具有这些独特优势的同时,存在着对载波频偏十分敏感的缺点。另一方面,OFDM在接收端一般采用相干检测,准确的信道状态信息必不可少,然而在高速移动场景下,传播路径复杂,信号衰落严重,信道己由传统的慢时变转变为快时变,因此,信道估计是OFDM系统需要解决的又一关键问题。基于此,本文研究分析了快时变信道条件下的OFDM系统频率同步与信道估计技术。论文首先介绍了OFDM系统模型和无线信道的仿真模型,在此基础上,分析推导了载波频偏对OFDM系统的影响,然后对基于循环前缀的频率同步技术和基于训练序列的频率同步技术做出了分析比较,研究了训练序列对频率同步性能的影响。其次,针对快时变信道的特点,分析了传统LS算法和MMSE算法的局限性,对基于基扩展模型的信道估计算法做了重点介绍,并给出了数据符号位置的信道插值算法。最后,考虑到频率同步与信道估计技术的内在联系,对比分析了传统联合估计算法和基于基扩展模型的联合估计算法的性能差异,并且详细推导了快时变信道条件下联合估计算法的理论界。仿真结果表明,基于训练序列的频率同步性能要明显好于基于循环前缀的频率同步性能,并且通过改变训练序列的种类、长度和插入位置可以进一步提高同步性能,另一方面,基于基扩展模型的信道估计算法以及频偏和信道联合估计算法在快时变信道条件下能够提供较为理想的估计性能。
[Abstract]:With the continuous expansion of the number of users and the emergence of a large number of multimedia services, the third generation of mobile communication systems can no longer meet the development needs of future mobile communications. Therefore, the 3GPP organization initiated the long-term evolution of 3G technology, (Long Term Evolution,. As one of the key technologies in the physical layer of LTE, LTE can not only provide high data transmission rate, but also can resist multipath fading and inter-symbol interference. However, while OFDM has these unique advantages, it has the disadvantage that it is very sensitive to carrier frequency offset. On the other hand, OFDM generally uses coherent detection at the receiver, and accurate channel state information is necessary. However, in high-speed mobile scenario, the propagation path is complex and the signal fading is serious. The channel has changed from traditional slow-time to fast-time-varying. Therefore, channel estimation is another key problem to be solved in OFDM system. Based on this, the frequency synchronization and channel estimation techniques for OFDM systems under fast time-varying channel conditions are studied and analyzed in this paper. Firstly, the OFDM system model and the simulation model of wireless channel are introduced. On the basis of this, the influence of carrier frequency offset on OFDM system is analyzed and deduced. Then the frequency synchronization technology based on cyclic prefix is analyzed and compared with the frequency synchronization technology based on training sequence, and the influence of training sequence on the performance of frequency synchronization is studied. Secondly, according to the characteristics of fast time-varying channel, the limitations of traditional LS algorithm and MMSE algorithm are analyzed, and the channel estimation algorithm based on the basis extension model is introduced emphatically, and the channel interpolation algorithm of data symbol position is given. Finally, considering the inherent relationship between frequency synchronization and channel estimation technology, the performance differences between the traditional joint estimation algorithm and the joint estimation algorithm based on the basis-extended model are compared and analyzed. The theory of joint estimation algorithm under fast time-varying channel conditions is deduced in detail. The simulation results show that the performance of frequency synchronization based on training sequence is obviously better than that based on cyclic prefix, and the synchronization performance can be further improved by changing the type, length and insertion position of training sequence. The channel estimation algorithm based on the basis-spread model and the joint frequency offset and channel estimation algorithms can provide ideal estimation performance under fast time-varying channel conditions.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.53

【共引文献】