【作者】 王艺霏；

【导师】 李立华；

【作者基本信息】 北京邮电大学 ， 通信与信息系统， 2013， 硕士

【摘要】 随着无线数据业务需求的迅速增长和新的空口技术不断引入,移动通信系统在支持高速多媒体业务传输以及实现无缝覆盖等方面不断面临新的挑战。为了满足IMT-Advanced (International Mobile Telecommunications-Advanced:国际移动通信增强系统)系统在带宽,峰值速率,通信吞吐量等方面的需求,多天线增强(Enhanced MIMO: Enhanced Multiple Input Multiple Output)技术、协作多点传输技术(CoMP:Coordinated Multi-point Tx/Rx)等作为关键技术被3GPP (3rd Generation Partnership Project:第三代合作伙伴计划)组织写入LTE-Advanced (Long Term Evolution-Advanced)协议中。在新一代移动通信系统的研究发展背景下,以实验室项目为支撑,本文进行了协作多点传输系统中多天线增强技术的研究,重点研究多用户MIMO的预编码技术和与相应预编码技术结合的功率分配方案。本文对协作多点传输系统进行了系统的研究和阐述,并进行了基本的分类。联合处理(JP:Joint Processing)和协作调度／波束成形(Coordinated Scheduling/Beamforming)为CoMP下行传输技术的两个基本分类。其中联合处理技术又可以具体分为联合传输技术(JT:Joint Transmission)及动态小区选择技术(DCS:Dynamic Cell Selection)。联合传输(JT)是联合处理技术两种方案的主要形式,也是intra-eNodeB的多点协作的主要传输方案。在多用户MIMO系统关于预编码技术的研究中,本文主要关注基于非码本的预编码技术,并从迫零(ZF:Zero Forcing)准则,最小均方误差(MMSE:Minimun Mean Square Error)准则和最大化信泄漏噪声比(SLNR:Signal to Leakage and Noise Ratio)准则考虑,研究了不同多用户MMO系统的预编码技术,重点研究了SLNR预编码技术,并进行了分析和比较。在SLNR预编码和功率分配结合方案的研究中,本文从提高通信传输可靠性和有效性的角度分别对目前已有的功率分配结合方案进行研究,并提出了以提高传输质量为出发点,以系统总均方误差之和最小为优化目标的SLNR预编码和功率分配结合的方案。通过构造拉格朗日算子,借助凸优化理论,本文提出采用迭代的操作方式获取功率分配因子,预编码矩阵以及接收译码矩阵的局部最优解,获得了误码率性能的提升。基于复杂度的考虑,又提出了次优的不进行迭代操作的SLNR预编码和功率分配结合算法。次优算法以降低部分误码率性能为代价,换取计算复杂度的降低,仍然保持在全局信噪比范围均优于等功率分配方案的性能优势。更多还原

【Abstract】 With the rapid growth of the demand for wireless data services and the introduction of new air interface technologies, the mobile communication system is facing new challenges to support high-speed multimedia services transmission and guarantee seamless coverage. In order to meet the requirement of IMT-Advanced (International Mobile Telecommunications-Advanced) system, including the demand for bandwidth, peak data rate and system throughput, Enhanced MIMO(Multiple Input Multiple Output) technology, CoMP (Coordinated Multi-point Tx/Rx) were written in LTE-Advanced (Long Term Evolution-Advanced) Protocol as a key technologies by3GPP (3rd Generation Partnership). With the rapid development of the next generation mobile communication system and support by laboratory projects, the thesis researches multi-antenna enhancement technology in coordinated multipoint transmission system and focuses on multi-user MEMO (MU-MLMO) precoding technologies and associated power allocation schemes.The thesis gives systematical research and basic classification of coordinated multipoint transmission system. The basic classifications of CoMP downlink transmission technology are joint processing (JP) and collaborative scheduling/beamforming (CS/CB). Joint processing technology includes joint transmission (JT) technology, and dynamic cell selection technology (DCS). Joint transmission is adopted as the main form of joint processing, also the main transmission scheme for multi-point collaboration in intra-eNodeB scenario.For precoding technology for MU-MIMO system, this thesis focuses on non codebook precoding technology. Based on zero forcing (ZF) criteria, the minimum mean square error (MMSE) criteria and maximun signal to leakage and noise ratio (SLNR) criteria, the thesis studies MU-MIMO precoding technologies and gives the analysis and comparison between several precoding schemes. And the main focus is on SLNR precoding technology.To the study on SLNR precoding and power allocation scheme, the thesis researches on the existing power allocation schemes for SLNR precoding from the view of communication transmission reliability and validity. Aiming at improving the transmission quality, minimizing the total system mean square error is used as of the optimization objective for SLNR precoding and power allocation scheme. By constructing the Lagrangian fuctiong and using the convex optimization theory, an optimal power allocation scheme with iterative operation is proposed. By this proposal, locally optimal solution for power allocation factor, precoding matrix and receiver decoding matrix are obtained ain each iterative procedure. This iterative power allocation scheme for SLNR precoding gives BER performance improvement. By considering of the complexity of power allocation algorithm, a suboptimal power allocation scheme for SLNR precoding is proposed. By trading off between BER performance and algorithm complexity, suboptimal power allocation scheme give reduction on computational complexcity with the cost of losing BER performance, still holding the advantage in all SNR regines when comparing the BER performance with equal power allocation algorithm.更多还原