【作者】 温沛霖；

【导师】 郝莉；

【作者基本信息】 西南交通大学 ， 通信与信息系统， 2012， 硕士

【摘要】 随着移动通信业务需求的持续增长,移动通信已经成为当前世界上技术发展最迅速的学科之一。多输入多输出技术(MIMO)作为新一代移动通信领域的关键技术之一,也是目前移动通信领域的研究热点,更优于单天线系统,MIMO技术能较好地提高信息传输的速率,改善信息传输的质量,提升通信容量。目前,多输入多输出技术已经广泛应用于宽带无线接入以及长期演进(LTE)等无线通信系统中。MIMO系统的性能提升与MIMO无线信道的传输特性关系较大。如果系统能够准确获得信道的状态信息,并且充分将其利用于自适应传输技术中,能显著提升通信系统的系统容量。。通信系统可以根据信道状态信息进行空时编码,预编码,自适应调制,功率控制等,提升通信质量。在高速移动的通信系统中,由于信道快变,当前时刻通过信道估计出来的信道状态信息不再适合指导未来时刻发射端发送下行数据。下一代蜂窝通信需要支持快达500km/h的超高移动速度。所以有必要使发射端准确获取未来的信道状态信息,或者直接减轻甚至克服对信道状态信息的需求,以显著改善系统性能。系统精确传输的关键在于准确的信道状态信息,由于信道预测技术能有效修正不准确的信道状态信息,是近年来MIMO增强技术的研究热点之一。系统通过参考符号对当前时刻的信道状态信息进行估计,并依据当前时刻的信道状态信息预测或者修正未来时刻的信道状态信息,从而提升MIMO系统传输的性能。随着高速铁路的快速发展,移动台的移动速度不断提高,移动传播环境越来越恶劣,某些传统的通信技术适用性不断受到挑战,酉空时编码技术不同于传统的空时编码技术,可以应用于未知信道状态信息的恶劣信道中,以克服系统无法正常获取CSI的困难。本文首先对MIMO信道预测与跟踪技术进行了研究。MIMO信道预测技术主要考察了几种常用的信道预测算法,并通过预编码技术对信道预测的性能进行了进一步验证。本文考察的算法包括基于MMSE和基于自适应滤波器的预测算法。基于MMSE的预测算法思想是信道的全局统计特性,基于自适应的预测算法思想是数据的局部平稳性,依据使用自适应滤波器的种类可以分为基于NLMS(归一化最小均方误差),RLS(递归最小二乘法)和Kalman这三种方法。最后,结合局部平稳性和全局统计特性,提出了引入预测样本迭代的信道预测策略,在预测深度加大的情况下有较好预测性能。本文还对酉空时编码技术进行了研究,其中包括传统的酉空时编码技术和差分酉空时编码技术,分别对两种酉空时编码技术在不同星座构成,不同多普勒扩展,不同检测方法,不同信道时变性等方面进行了充分的考察。最后,结合信道的时变特性,提出了一种适用于快速时变信道情景下,基于修正因子的差分酉空时检测技术,相对于传统的差分酉空时解码,有一定程度的性能提升。更多还原

【Abstract】 With the growing demand for mobile communications business from society, mobile communications is one of the fastest growing disciplines currently. Multiple input multiple output (MIMO) is a key technology in the new generation mobile communication, which is also a currently hot research field of mobile communications. Compared with the single antenna system, MIMO systems can significantly improve the information rate, the quality of information transmission, and the communication capacity. Currently, MIMO technology has been widely used in broadband wireless access, such as long-term evolution (LTE) wireless communication system.The performance of MIMO systems greatly dependents on the transmission characteristics of MIMO wireless channel. Accurate channel state information (CSI) plays a vital role improving the system capacity. Communication system uses the channel state information to achieve space-time coding, precoding, adaptive modulation, power control and so on.In the mobile communication system, the CSI is obtained from channel reciprocity and feedback. The CSI in the present moment is no longer suitable for the system to send downlink data in the future, ie, the time varying characteristic lead to the channel state information out of date. It is difficult to obtain the channel state information in the context of fast fading channel. The next generation wireless communications need to support ultra-high mobile speed up to500km/h. Therefore, it is necessary for the system to obtain accurate channel state information in the future, or directly reduce or even overcome the channel state information, in order to significantly improve system performance.The channel prediction technique can effectively modify the inaccurate channel state information, and is one of the popular techniques in recent years. The system estimate the channel state information of the current moment by using reference symbols and use the channel state information to predict or modify the channel state information in the future.With the continuous development of high-speed railway, the mobile communication environment is worsening, the applicability of some traditional communication technologies is constantly being challenged. Different from the traditional space-time coding techniques, the unitary space-time coding techniques, can be applied in a very bad channel with unknown channel state information, in order to overcome the system can not normally obtain CSI.First, the MIMO channel prediction techniques in researched. Several common channel prediction algorithms are investigated, including MMSE and adaptive filter. The performance of precoding by channel prediction is studied as further verification.. The MMSE prediction algorithm is based on channel global statistical properties. The adaptive filter algorithm is based on partial stability of data, according to the structure, adaptive filter can be divided into these three methods:NLMS, RLS and Kalman. Finally, with the combination of local stability and global statistical properties, new prediction strategy is proposed.Then, the unitary space-time modulation techniques are studied, including traditional unitary space-time coding techniques and differential unitary space-time coding techniques. Further, various simulations in different constellations, Doppler expansion, demodulation method and time varying channel is analyzed. Finally, a new differential demodulation method is proposed base on channel prediction. Compared to conventional demodulation method, the demodulation performance is increased.更多还原