【作者】 谢显中；

【导师】 王新梅；

【作者基本信息】 西安电子科技大学 ， 通信与信息系统， 2000， 博士

【摘要】 空时处理技术能很好的结合单时间处理和单空间处理各自的特点，这是近年来信号处理领域关注的热点。由于空时处理技术的优势，移动通信系统正在采用空时处理技术来抵消系统干扰、改善处理增益、增加小区覆盖范围、减少发送功率、提高系统容量。本论文研究空时处理技术在移动通信系统应用中的空时编码技术和接收机设计问题。空时编码是指当用多天线分集发送时的信号设计技术，它是最大化传输空时分集增益的技术，由于移动通信复杂的信道特点，空时编码技术一般和信道编码技术结合在一起，从而我们所指的空时编码技术就是这种结合技术。 本论文首先讨论了空时信道、空时信号模型，综述了空时处理技术、空时编码技术的现状和前景。然后研究了神经网络在空时编码中应用、分层空时编码的译码及其性能仿真、Turbo码与空时码的结合及其性能、空时编码接收机设计、空时处理接收机设计等方面。主要贡献包括： ● 讨论空时信道、空时信号模型，给出了一种空时时变矢量信道模型及其分 类，其次系统地分析了空时信号的一般模型和离散化的空时模型。对已有 的空时处理技术成果从理论发展到实际实现两方面进行全面的总结。 ● 对空时编码技术按照作者的理解进行界定，然后讨论了空时编码的基本思 想和原理，从信息论角度分析了天线阵分集传输系统的容量，对分层空时 码（BLAST）、Trellis空时码（STCM）和分组空时码（BSTC）编译码技术、 特点、性能等进行全面分析。 ● 借助信道编码的Trellis图和神经网络的平均场退火算法，给出了一种新的 Hopfeild神经网络译码方案，其性能接近最大似然译码。同时由于其能量 函数阶仅为2，这样设计的Hopfeild神经网络结构简单，收敛速度快，容 易用数字硬件电路实现。 ● 针对对角分层空时编码方案，给出了具有线性复杂度的迫零判决反馈均衡 器分层空时接收与译码的算法，仿真结果显示了良好的性能和更高的频谱 利用率。 ● 首先讨论了Turbo码调制技术及其文献中的缺陷，然后研究了Turbo码与 空时编码中的STCM结合问题，通过仿真实验与比较，证明了基于Turbo 码的STCM系统具有性能优势，能满足室内和低移动性环境的高速无线 数据通信要求。 ● 采用群多用户检测和BSTC解码的线性检测相结合的两级检测技术设计卫 西安电子科技大学博士学位论文 出多用户的BSTC系统接收机，仿真结果证明了良好的性能，并改进了部 分文献中存在的一些问题。 ．给出了一种上行多用户空时信道模型，基于该信道模型导出了一种空时滤 波和多用户CCI、ISI消除器的空时接收机设计结构。讨论了空时接收机 中的结合DOA估计的联合信道估方法，并针对提出的接收机结构的工程 实现进行改进，最后仿真实验验证了空时接收机性能。更多还原

【Abstract】 Space-time processing technique, which has strong points of both time processing and space processing, is considered as one of the most important field of signal processing. Nowadays, mobile communications are rapidly growing around the world and a number of emerging techniques are employed for canceling interference, improving gain, extending coverage, decreasing transmit power and increasing capacity in mobile communication systems. Among these techniques the most exciting and potentially one is the use of space-time processing. This thesis investigates some problems of space-time processing that applies to mobile communications with emphasis on space-time code techniques and receiver design. Space-time code is a signal design technique for transmit diversity with antenna arrays, and maximizes the transmit diversity gain. Because of multi-path and time-varying propagation environment for mobile communication systems space-time code usually combines with channel coding technique. In the thesis the generalized space-time code technique, which combines with channel coding, is treated. Firstly, this thesis discusses mobile communication channel and space-time signal model. Secondly, current studies and future prospects of both space-time processing and space-time code are given. Lastly, we research space-time code technique and receiver design including application of neural networks to space-time code, a transmitter and receiver model employing layered space-time and its performance simulation, the turbo STCM that combine of turbo code and space-time code and its performance. receiver design of multiuser space-time code systems, receiver design of mobile communication system employing space-time processing. The main results are as follows: ? Space-time channel and space-time signal are studied. A kind of model of space-time vector channel is given, and space-time signal is analyzed systemly. Current studies and future prospects of space-time processing techniques with the emphasis from theory to application are discussed. ? According to author抯 opinion the definition of space-time coding is introduced. Then the principle of space-time coding and the capacity of antenna arrays transmit diversity system are discussed from information theory aspect. Also, the strong points and performances of BLAST, STCM and BSTC system are analyzed. ? A new decoding scheme is given based on Hopfield neural networks. The order of the energy function of the Hopfield network is reduced to 2 by means of Trellis diagram and mean field annealing algorithm. It is important improvement compared with previous work published and is implemented easily by hardware. The simulation result shows the decoding performance approaches the ideal MLD. ? A transmitter and receiver model employing layered space-time codes is discussed. Further, encoding methods of layered space-time codes are derived. Also the decoding and receiving scheme using linear Zero-forcing Iv decision-feedback equalizer with linear processing complexity is studied. Lastly. simulation results show good performance and higher spectral efficiency. ? At first, we discuss turbo code modulation technique, and some problems in previous work published are analyzed. Then the turbo STCM that combines turbo code and STCM is driven. We investigate transmit and decode schemes of the turbo STCM. The simulation results compared with更多还原