【作者】 石海；

【导师】 陈惠民；

【作者基本信息】 上海大学 ， 通信与信息系统， 2008， 博士

【摘要】 随着蜂窝移动通信、因特网和多媒体业务的发展,世界范围内无线通信系统的容量需求迅速增长。然而,由于目前可利用的频谱已十分拥挤,以及移动和便携终端有限的发射功率,寻求更高效的信号传输技术已成为数字无线通信系统研究领域最活跃的研究课题之一。在追求高效性的同时,无线数字通信系统的高可靠性始终也是从事无线通信系统研究的学者和工程师最关注和最活跃的研究领域。近年来的研究成果已经证实:采用Turbo编码和最大似然译码方法构造的系统是目前系统性能逼近香农限的最佳方法;利用多输入多输出(Multiple Input MultipleOutput,MIMO)空间复用技术能够显著提高通信系统的频谱利用率。近年来,为了充分发挥两者的优势,开发具有大容量、高可靠性的Turbo MIMO宽带无线通信系统,国内外众多研究机构在该系统的研究中倾注了大量人力和物力,并在理论上取得了令人瞩目的成绩。然而,由于Turbo MIMO系统自身,以及无线信道的复杂性,Turbo MIMO系统的研究仍然存在着众多有待解决的问题。为了进一步加快开发可行的、大容量、高可靠性的Turbo MIMO通信系统的研究步伐,本文在及时跟踪相应研究领域国内外发展动向的基础上,针对Turbo MIMO通信系统的下列四个问题进行了深入的研究,并取得了一些新的研究成果。首先,针对目前基于混合型序贯蒙特卡洛(Sequential Monte Carlo,SMC)算法的Turbo MIMO接收系统无法适应无线信道随机变化的缺陷,本文引入了一种新的动态界定参数d,并在此基础上结合动态界定参数d的设定,分别提出了直接随机化和系统随机化动态混合型SMC Turbo MIMO接收机空时译码算法。其中,直接随机化动态混合型SMC方案通过在迭代过程中直接随机地修正界定参数d,以寻求在统计意义上提高系统的整体性能;而系统随机化动态混合型SMC方案则间接地利用MIMO信道的变化来修正界定参数d,以寻求提高算法跟踪信道变化的能力。仿真验证,两种随机化方案在不增加算法复杂度开销的情况下,都能够有效地提高原有混合型SMC方案的性能。其次,针对富散射信道环境下MIMO信道呈瑞利衰落的特性,以及Turbo系统在低码速时存在着严重的载波相位偏移问题,本文采用导频符号辅助调制(PilotSymbol Assisted Modulation,PSAM)技术来估计瑞利衰落信道,提出了一种在瑞利衰落信道和加性白高斯噪声(Additive White Gaussian Noise,AWGN)干扰的共同作用下,基于PSAM技术的联合Turbo译码和载波相偏恢复方法,并通过仿真验证了该方法能以较小的系统开销,对载波相位进行较好的恢复。其三,针对目前采用分级横向自适应滤波算法,由于其子滤波器工作在前向预测状态而非滤波状态,致使对信道均衡误差较大的缺陷,本文提出了一种能使所有子滤波器均工作在滤波状态的改进型分级横向自适应滤波算法。并在此基础上,分别设计了基于LMS和基于RLS的改进型分级横向自适应滤波的信道均衡器。研究结果表明,在相同条件下两种改进型信道均衡器不仅仍能保持快速均衡的特性,而且能比目前分级横向自适应滤波算法的均衡误差至少减少一个数量级。最后,为了寻求垂直贝尔实验室分层空时编码(Vertical Bell Labs LayeredSpace-Time,V-BLAST)MIMO系统在室外宏蜂窝非富散射环境中的应用,本文根据Ali Abdi等人提出的宏蜂窝非富散射信道模型,着重研究了入射波到达角(AOA,Angel of Arrival)非均匀分布时,对V-BLAST MIMO移动系统性能的影响。研究发现系统误码特性随移动台入射波AOA的角度扩散减小而下降,而基站入射波的波束宽度对系统性能影响较小。该研究成果为评估入射波AOA非均匀分布对移动台性能影响提供了理论依据,也为MIMO系统在室外宏蜂窝非富散射环境,以及空时相关信道条件下的研究,提供了一定的可供参考的方法。更多还原

【Abstract】 With the development of cellular wireless communication, Internet and multimedia services, the demand for capacity of worldwide wireless communication systems is growing at a very rapid pace. Dou to the severe limitation of available radio spectrum and the transmitter power provided by mobile or other portable terminal, seeking the more effective signal transmission techniques has become one of the most active research subjects in digital wireless communication system. In the progress of seeking effective techniques, the link reliability of wireless system is always one of the most attentive and active problems for researchers and electronic engineers who are working in the wireless communication systems.The research results in recent years have approved that the performance of the system with Turbo code and max likelihood decoding could approach to Shannon limit and the spectrum efficiency of wireless system could remarkably increase by using MIMO spatial multiplexing technology. To fully utilize these technology advantages and develop broadband wireless Turbo MIMO system with large capacity and high reliability, most research organizations throw all their energy into this work and have obtained some inspiring scientific achievements. There are, however, a lot of problems required for Turbo MIMO system, because of the difficulty of system itself and the complicacy of wireless channels.To expedite the course of developing feasible Turbo MIMO system with large capacity and high reliability, the four research projects, as follows, on Turbo MIMO system have been done based on recent research results in all over the world. There are some new results obtained from research.Firstly, for overcoming the disadvantage of original combined SMC (Sequential Monte Carlo) Turbo MIMO receiver that could not adapt random variety of wireless channels, two new combined SMC Turbo MIMO space-time decoding algorithms with dynamic boundary parameter d are proposed. They are called direct randomization dynamic combined SMC Turbo MIMO algorithm and system randomization SMC Turbo MIMO algorithm, respectively. In the direct randomization method, boundary parameter d is randomly modified, in order to improve the performance of system in sense of statistics. In the system randomization method, boundary parameter d is modified by utilizing indirectly the information from channels, to improve the ability that algorithm follows channel variety. The computer simulation results show that the space-time decoding performance could be enhanced by the two new algorithms without the burden of complexity related to the original combined scheme.Secondly, for the problem of the carrier phase recovery in Turbo MIMO system with low speed transmission over Rayleigh fading channels, a new carrier phase recovery scheme suited for Turbo systems in Rayleigh fading channels with AWGN is proposed. It is called joint carrier phase recovery and Turbo decoding based PSAM (Pilot Symbol Assisted Modulation) technology. This scheme involves the estimation with PSAM for Rayleigh fading channel and the joint phase recovery and turbo-decoder for AWGN. Simulation results show that the proposed method with small system expense can significantly improve the performance of carrier phase recovery for Turbo systems in Rayleigh fading channels with AWGN.Thirdly, for the disadvantage of original hierarchical transversal adaptive filtering algorithm with larger error in channel equalization because the filter operates at the state of prediction not at the state filtering, a modified hierarchical transversal adaptive filter is proposed, in witch each subfilter could operate at the state of filtering. Based on the new scheme, the modified hierarchical transversal adaptive filtering algorithms based LMS (Least Mean Square) and RLS (Recursive Least Square) are proposed, respectively. The results of simulation indicate that, under same conditions, the modified HLMS and HRLS algorithms can not only maintain the feature of speeding up convergence rate, but also obtain at least one order of magnitude improvement for the mean square error of channel equalization related to the original scheme.Lastly, to seek the application of V-BLAST (Vertical Bell Labs Layered Space-Time) MIMO system under macro-cell environments without rich-scattering, the performance of V-BLAST system with the nonuniform distribution of AOA (Angel of Arrival) of incident radio has been researched by the macro-cell MIMO channel model proposed by Ali Abdi. The research finds that BER for mobile station will increase with the angel diffusion decrease for AOA of incident radio, but the performance of base station is hardly affected by the beamwidth of incident radio. The results provide scientific basis for evaluating the performance of mobile station with the nonuniform distribution of AOA of incident radio, moreover, it could supply some references to the research of MIMO system under the macro-cell environments without rich-scattering and space-time correlative channels.更多还原