【作者】 魏旻；

【导师】 龚耀寰；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2007， 博士

【摘要】 从20世纪末到21世纪初，无线通信技术在全世界范围内以前所未有的速度向前发展，人们对通过无线方式随时随地接入因特网获取信息以及在手机上实现更多多媒体业务的需求也显得日益迫切，这就为新一代无线通信系统的产生打下了一个坚实的市场基础，同时也吸引了全世界范围内众多科研人员投入到新的无线通信技术的研究中。MIMO技术和OFDM技术是未来无线通信系统中的2个重要的技术，而将MIMO技术和OFDM技术结合在一起形成的MIMO-OFDM系统能够使频谱效率得以成倍提高，也能使信道容量得到提高；同时大大降低在多径环境下的接收机复杂度，因此成为了未来无线通信系统中的核心解决方案。为了在未来无线通信系统中充分发挥MIMO-OFDM系统的优越性能，需要对MIMO-OFDM系统中的编解码、系统方案、信道分析、信道估计及实现等方面进行更深入的研究。本文对MIMO-OFDM系统的方案进行了深入的研究，提出了具有更低运算量更好误码率性能的TR-STBC-OFDM系统方案；在此基础上利用几何方法推导出了多信道间相关系数的封闭解表达式，分析了相关性对MIMO-OFDM系统的影响；为了更好地提高TR-STBC-OFDM系统的频谱利用率及性能，对基于二阶统计特性的信道盲估计技术进行了研究，提出了在接收天线数少于或等于发射天线数情况下的信道估计方法，克服了子空间方法及线性预测方法在这方面的不足。本文的主要工作包括：一、针对传统的MIMO-OFDM系统，提出了发射天线数为2和3、发射信号为实信号及复信号下的一种新的解码方法。该解码方法将信道矩阵与编码矩阵结合在一起，使得在进行空时解码的过程中，不仅完成了空时解码，而且还可同时将复信号映射为实信号，为接收机后续工作的进行提供了更便利的条件，有助于简化接收机的设计。二、将TR-STBC编码技术与OFDM技术结合起来，提出了2种新的MIMO-OFDM系统方案。这两种方案一方面利用了TR-STBC编码技术的优势，使得在频率选择性信道下，较传统的STBC编码技术能在不损失分集增益的前提下更方便地将多发天线转换为单发天线；另一方面通过采用OFDM技术，能够将频率选择性信道方便地转换为一组频率平坦性信道，大大简化了接收机的结构。除了以上提到的优点外，本文所提出的2种TR-STBC-OFDM系统方案还具有如下优点：(1)、本文提出的TR-STBC-OFDM系统方案1在发射端首先进行1次IFFT运算完成OFDM调制后再进行TR-STBC编码；在接收端先进行TR-STBC解码再进行一次FFT运算完成OFDM解调。通过这样的操作流程，TR-STBC-OFDM系统方案1较传统的MIMO-OFDM系统减少了运算量，同时通过仿真实验可看到本文提出的TR-STBC-OFDM系统方案1具有更好的误码率性能。(2)、本文提出的TR-STBC-OFDM系统方案2在发射端进行IFFT运算的次数与传统的MIMO-OFDM系统相同，但减少了进行IFFT运算的点数，在接收端进行FFT运算的次数与发射天线数相同，但每个FFT运算的点数都得到了减少。通过这样的操作，从另一个角度减少了系统的运算量，同样通过仿真实验可看到本文提出的TR-STBC-OFDM系统方案2具有更好的误码率性能。三、建立了MIMO-OFDM系统背景下微小区中无线信道间相关系数的封闭解表达式。该封闭解表达式是基于通用的微小区椭圆信道模型，利用几何方法进行推导而得到。仿真结果指出由于在微小区中具有丰富的散射体，所以不论是在基站(BS)端附近还是在移动台(MS)端附近，都具有较大的角度扩展，在较小天线阵元间距时无线信道间亦能达到较小的相关性，因此本文提出在微小区中可以设置较小的收/发端天线阵阵元间距，达到一个较小的相关系数。该结论能够帮助设定3GPP的SCM模型的相关系统参数，更好地为系统方案及算法的验证服务。四、针对本文提出的TR-STBC-OFDM系统中下行链路的情况，提出了基于二阶统计特性的信道盲估计技术在接收天线数等于或少于发射天线数情况下进行信道盲估计的方法。该方法充分利用了TR-STBC-OFDM系统中TR-STBC技术带来的新特性，利用矩阵变换工具，方便地将接收天线数少于或等于发射天线数的系统转化为接收天线数大于发射天线数的系统，从而满足了基于二阶统计特性的子空间法、线性预测方法等信道盲估计方法要求接收天线数大于发射天线数的条件。通过信道盲估计技术的研究也可以看到，本论文中所提出的TR-STBC-OFDM系统方案较传统的MIMO-OFDM系统方案不仅仅能够减少运算量提高误码率性能，而且能够为基于二阶统计特性的信道盲估计方法在下行链路中的使用提供强有力的支持。更多还原

【Abstract】 From the end of the 20th century to the beginning of the 21st century, the wireless communication technology develops forward within the whole world range with hitherto unknown speed. People also appears gradually urgent to the need for gaining information and realizing more medium business on mobile telephone at any time and any place. It establishes solid market base for the generation of the new generation wireless communication system, at the same time it has attracted a lot of scientific research personnel.MIMO and OFDM are two important techniques in future wireless communication system. MIMO-OFDM system, which combines MIMO with OFDM, can improve spectrum efficiency, the channel capacity and simplify the receiver structure under the multipath environment. So MIMO-OFDM is a key scheme for the future wireless communication system. Based on the importance of MIMO-OFDM, the encoding/decoding scheme, system scheme, channel analysis, channel estimation and system implementation are studied by more research personnel.In this thesis, the author deeply studied the MIMO-OFDM scheme. The TR-STBC-OFDM with lower compute complexity and better BER performance is brought forward. The channel correlation coefficient in microcellular has been derived by making use of geometry method. And the blind channel estimation algorithms which is based on second order statistics are studied under the TR-STBC-OFDM system.The main contributions of the thesis include:1. A novel decoding scheme for MIMO-OFDM system is brought forward. In this scheme, channel matrix is combined with encoding matrix to form new decoding matrix. Through this decoding matrix the complex signal is converted conveniently into real signal, at the same time the space-time decoding is completed.2. Combining TR-STBC technology with OFDM technology to form two novel TR-STBC-OFDM system schemes. In these two novel schemes, TR-STBC is adopted to convert multiple transmitting antennas into single transmitting antenna; OFDM is adopted to convert frequency-selective channel into a group of frequency-flat channels.1) TR-STBC-OFDM system scheme 1 changes the position of OFDM modulator and space-time encoder, so it only need one IFFT at transmitter and one FFT at receiver. Through this way, TR-STBC-OFDM system scheme 1 cuts down the compute complexity. Simulation results indicate that TR-STBC-OFDM system scheme 1 has better BER performace than tranditional MIMO-OFDM system.2) TR-STBC-OFDM system scheme 2 also changes the position of OFDM modulator and space-time encoder. At transmitter, the times of IFFT is the same with the tranditional MIMO-OFDM system but it decreace the points of the IFFT; at receiver the times of FFT is equal with the transmitting antennas and the points of the FFT are also decreased. Through this way, TR-STBC-OFDM system scheme 2 cuts down the compute complexity. Simulation results indicate that TR-STBC-OFDM system scheme 2 also has better BER performace than tranditional MIMO-OFDM system.3. Making use of geometry method to analyze the correlation of the wireless channel and the close expression of the correlation coefficient is derived. Here, the ellipse model is adopted, because it can better describe the microcellular which has rich scatters. These results can effectively help the design of the MIMO-OFDM system and the analysis of the correlation in MIMO-OFDM system.4. The blind channel estimation which is based on the second order statistics has been studied. The method for the situation that the receiving antennas are less or equal the transmitting antennas in downlink is brought forward. Under the TR-STBC-OFDM system, through the new characteristic of TR-STBC, 2 transmiting antennas 2 receiving antennas system can be converted into 2 transmiting antennas 4 receiving antennas system, and 3 transmiting antennas 1 receiving antenna system can be converted into 3 transmiting antennas 4 receiving antennas system. Through this way, the condition that the receiving antennas must be more than transmitting antennas in blind channel estimation algorithm which is based on the second order statistics is satisfied One hand this result can help the application of the blind channel estimation algorithm in downlink, at another hand, it indicates the TR-STBC-OFDM system brought forward by this thesis can powerfully support the application of the blind channel estimation which is based on second order statistics.更多还原