大规模MIMO系统能效优化研究
发布时间：2018-12-28 13:32
【摘要】：通过使用大规模天线阵完成数据传输,大规模多输入多输出(Multiple Input Multiple Output, MIMO)系统可以将波束能量集中到很小的空间区域,从而节省发射功率并降低对非目标用户的干扰,因此可以实现比传统MIMO系统更高的能量效率,是重要的5G (5 Generation,第五代移动通信系统)候选技术。随着全球变暖的不断加剧,能效优化成为科研和工程实现的一个重要的关注点,在大规模MIMO系统中,由于大量射频链路消耗的功率较大,使用所有天线完成数据传输会造成能效损失,另外,系统复杂化为能效进一步优化提供了空间,因此大规模MIMO的能效优化问题近年来得到了广泛的研究。本文重点关注考虑射频链路功耗的天线选择问题以及天线数、用户数和发射功率联合优化问题,本文的主要工作有：1.对低复杂度天线选择算法进行了研究。针对大规模MIMO系统天线数目较多,天线选择复杂度较高的问题,首先介绍了一种以最小平均误差作为目标函数的上行天线选择算法,该算法可以转化为一个稀疏近似问题,并通过现有成熟的稀疏近似方法进行求解。基于稀疏近似的方法,提出了一种适用于大规模MIMO多用户系统的天线选择算法,该方法利用大规模天线阵所能获得的信号分集,以信道矩阵作为测量字典,通过稀疏近似方法求解低相关性的天线集。该方法考虑天线相关性,在大规模MIMO系统中性能优于不考虑天线选择的低复杂度天线选择算法,仿真结果表明,天线相关性越强,该算法的性能优势越明显。2.对迫零(Zero Forcing, ZF)预编码方式下的天线数、用户数和发射功率联合能效优化进行了研究。通过对ZF预编码方式下遍历能效的分析,结合分式优化算法,给出了多用户场景下的天线数、发射功率和用户数的联合优化方法及其退化形式。在优化过程中,发射功率和天线数优化问题均可转化为一种可以快速求解的形式,且满足相互迭代的条件；用户数优化问题也可以通过二分法来求解；三者的联合优化也可根据由表达式得出的性质进行简化。仿真结果表明,该算法在相关和非相关信道下均可实现近似最优的能效。3.对最大比发射(Maximum Ratio Transmission, MRT)预编码方式下的天线数、发射功率优化算法进行研究。根据ZF预编码方式下的联合优化方法的思路,对MRT预编码方式下的遍历能效进行推导,给出了MRT预编码方式下应用该方法的相关证明及方法的具体形式。本文对多用户系统和单用户系统两种系统模型均进行了研究,对于单用户系统,考虑了信道状态信息(Channel State Information, CSI)已知和未知两种场景。
[Abstract]:By using large scale antenna array to complete data transmission, large scale multi-input and multi-output (Multiple Input Multiple Output, MIMO) system can concentrate beam energy into a small space area, thus saving transmission power and reducing interference to non-target users. Therefore, it is possible to achieve higher energy efficiency than traditional MIMO systems and is an important 5G (5 Generation, fifth generation mobile communication system) candidate technology. With the increasing global warming, energy efficiency optimization has become an important concern in scientific research and engineering implementation. In large-scale MIMO systems, a large number of RF links consume a large amount of power. Energy efficiency loss can be caused by using all antennas to complete data transmission. In addition, system complexity provides space for further optimization of energy efficiency. Therefore, energy efficiency optimization of large-scale MIMO has been widely studied in recent years. This paper focuses on the antenna selection problem considering RF link power consumption and the joint optimization of antenna number, number of users and transmit power. The main work of this paper is as follows: 1. A low complexity antenna selection algorithm is studied. Aiming at the large number of antennas in large scale MIMO systems and the high complexity of antenna selection, an uplink antenna selection algorithm with minimum average error as the objective function is introduced, which can be transformed into a sparse approximation problem. It is solved by the existing sparse approximation method. Based on the sparse approximation method, an antenna selection algorithm for large-scale MIMO multiuser systems is proposed. The channel matrix is used as the measurement dictionary, and the signal diversity can be obtained by large scale antenna array. The sparse approximation method is used to solve the low correlation antenna set. This method takes antenna correlation into account, and its performance is better than the low complexity antenna selection algorithm without antenna selection in large-scale MIMO systems. The simulation results show that the stronger the antenna correlation is, the more obvious the performance advantage of the algorithm is. 2. The optimization of antenna number, number of users and transmit power under zero forcing (Zero Forcing, ZF) precoding is studied. Based on the analysis of ergodic energy efficiency in ZF precoding and the fractional optimization algorithm, the joint optimization method of antenna number, transmit power and number of users in multi-user scenarios and its degenerate form are presented. In the process of optimization, the optimization problem of transmit power and antenna number can be transformed into a form that can be solved quickly, and the condition of iteration can be satisfied, and the optimization problem of number of users can also be solved by dichotomy. The joint optimization of the three can also be simplified according to the properties obtained from the expression. Simulation results show that the proposed algorithm can achieve approximately optimal energy efficiency in both correlated and uncorrelated channels. The optimization algorithm of antenna number and transmit power in (Maximum Ratio Transmission, MRT) precoding mode is studied. According to the idea of joint optimization method in ZF precoding mode, the ergodic energy efficiency under MRT precoding mode is deduced, and the relevant proof and the concrete form of the method applied in MRT precoding mode are given. In this paper, two kinds of system models, multiuser system and single user system, are studied. For single user system, the known and unknown scenarios of channel state information (Channel State Information, CSI) are considered.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN919.3
本文编号：2394010
[Abstract]:By using large scale antenna array to complete data transmission, large scale multi-input and multi-output (Multiple Input Multiple Output, MIMO) system can concentrate beam energy into a small space area, thus saving transmission power and reducing interference to non-target users. Therefore, it is possible to achieve higher energy efficiency than traditional MIMO systems and is an important 5G (5 Generation, fifth generation mobile communication system) candidate technology. With the increasing global warming, energy efficiency optimization has become an important concern in scientific research and engineering implementation. In large-scale MIMO systems, a large number of RF links consume a large amount of power. Energy efficiency loss can be caused by using all antennas to complete data transmission. In addition, system complexity provides space for further optimization of energy efficiency. Therefore, energy efficiency optimization of large-scale MIMO has been widely studied in recent years. This paper focuses on the antenna selection problem considering RF link power consumption and the joint optimization of antenna number, number of users and transmit power. The main work of this paper is as follows: 1. A low complexity antenna selection algorithm is studied. Aiming at the large number of antennas in large scale MIMO systems and the high complexity of antenna selection, an uplink antenna selection algorithm with minimum average error as the objective function is introduced, which can be transformed into a sparse approximation problem. It is solved by the existing sparse approximation method. Based on the sparse approximation method, an antenna selection algorithm for large-scale MIMO multiuser systems is proposed. The channel matrix is used as the measurement dictionary, and the signal diversity can be obtained by large scale antenna array. The sparse approximation method is used to solve the low correlation antenna set. This method takes antenna correlation into account, and its performance is better than the low complexity antenna selection algorithm without antenna selection in large-scale MIMO systems. The simulation results show that the stronger the antenna correlation is, the more obvious the performance advantage of the algorithm is. 2. The optimization of antenna number, number of users and transmit power under zero forcing (Zero Forcing, ZF) precoding is studied. Based on the analysis of ergodic energy efficiency in ZF precoding and the fractional optimization algorithm, the joint optimization method of antenna number, transmit power and number of users in multi-user scenarios and its degenerate form are presented. In the process of optimization, the optimization problem of transmit power and antenna number can be transformed into a form that can be solved quickly, and the condition of iteration can be satisfied, and the optimization problem of number of users can also be solved by dichotomy. The joint optimization of the three can also be simplified according to the properties obtained from the expression. Simulation results show that the proposed algorithm can achieve approximately optimal energy efficiency in both correlated and uncorrelated channels. The optimization algorithm of antenna number and transmit power in (Maximum Ratio Transmission, MRT) precoding mode is studied. According to the idea of joint optimization method in ZF precoding mode, the ergodic energy efficiency under MRT precoding mode is deduced, and the relevant proof and the concrete form of the method applied in MRT precoding mode are given. In this paper, two kinds of system models, multiuser system and single user system, are studied. For single user system, the known and unknown scenarios of channel state information (Channel State Information, CSI) are considered.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN919.3
【相似文献】
相关期刊论文 前2条
1 刘文龙;裴莹莹;金明录;;BPSK通信系统的部分最优MIMO检测算法[J];信号处理;2013年10期
2 ;[J];;年期
相关重要报纸文章 前1条
1 特约撰稿人 奇琳;B3G的多天线技术[N];通信产业报;2006年
相关硕士学位论文 前2条
1 寇鹏;大规模MIMO系统能效优化研究[D];西安电子科技大学;2014年
2 马琰;MIMO-Y信道的多天线冗余处理算法[D];西安电子科技大学;2013年
，本文编号：2394010
本文链接：https://www.wllwen.com/kejilunwen/wltx/2394010.html
