5G系统中非正交多址接入技术的研究

发布时间：2018-10-22 07:18

【摘要】：第五代移动通信(5G)不仅要实现数据流量的爆发式增长,还要解决大规模连接和低延迟等问题。为了应对这些挑战,NOMA(Non-Orthogonal Multiple Access)可以利用功率域自由度,提高频谱利用率,成为5G的关键技术之一。NOMA允许多用户以不同发射功率共享频谱资源,并采用串行干扰消除接收机实现多用户检测。但是,一方面由于NOMA仅根据信号与干扰加噪声比这个特征区分不同用户的信号,接收机的性能严重依赖叠加传输的用户选择和功率分配方案。另一方面,波束成形技术可以抑制干扰并提高数据速率,但受信道状态信息(Channel State Information,CSI)估计误差的影响,NOMA的波束成形设计稳定性较差。针对NOMA接收机性能严重依赖用户匹配和功率分配方案的问题,本文研究了多载波NOMA系统中用户分组和功率分配算法。具体的,在总发射功率的约束下,研究多载波NOMA下行链路的加权和速率性能的最优化问题。由于子载波分配的组合特性,本文分两步解决所建立的非线性规划问题。首先,提出用户动态分组方案。通过最大化所有匹配用户的信道增益差异,所提方案可以提高系统吞吐量。然后,本文采用WMMSE算法来解决非凸的功率分配问题,以低复杂度的迭代优化可获得局部最优解,通过仿真验证了该算法的收敛性和有效性。针对NOMA波束成形下系统性能稳定性受CSI估计误差影响的问题,本文研究单载波NOMA中多输入单输出信道下的稳健波束成形设计。具体的,考虑两种类型的CSI误差:二阶有界CSI误差和二阶无界CSI误差。在椭圆区域限制的有界CSI误差模型中,本文提出发射功率约束下的用户无中断译码的目标速率最大化问题。为了解决这个问题,首先将其分解为一系列求功耗最小的优化问题。利用凸优化理论,可以得到稳定的波束成形向量,在存在CSI误差时仍获得较高的速率性能。在二阶无界CSI误差模型中,本文研究满足用户中断概率的服务质量下最小化发送功率的优化问题。中断概率的约束使原问题难以处理,本文通过解决这个问题,获得可以满足中断概率要求的稳健波束成形向量。仿真结果验证了所提算法的稳健性和有效性。
[Abstract]:The fifth generation mobile communication (5G) is not only to achieve explosive growth of data traffic, but also to solve the problems of large-scale connection and low latency. In order to meet these challenges, NOMA (Non-Orthogonal Multiple Access) can make use of the degree of freedom in power domain to improve spectral efficiency and become one of the key technologies of 5G. NOMA allows multiple users to share spectrum resources with different transmit power. And the serial interference cancellation receiver is used to realize multi-user detection. But on the one hand the performance of the receiver depends heavily on the superimposed transmission user selection and power allocation scheme because the NOMA only distinguishes the signals of different users according to the characteristic of the ratio of signal to interference plus noise. On the other hand, beamforming technology can suppress interference and improve data rate, but the stability of NOMA beamforming design is poor due to the estimation error of channel state information (Channel State Information,CSI). In order to solve the problem that the performance of NOMA receiver depends heavily on user matching and power allocation scheme, this paper studies the user packet and power allocation algorithm in multicarrier NOMA system. Specifically, under the constraint of total transmit power, the weighting and rate performance optimization of multicarrier NOMA downlink is studied. Due to the combination of subcarriers, the nonlinear programming problem is solved in two steps. First, a user dynamic grouping scheme is proposed. By maximizing the channel gain difference of all matched users, the proposed scheme can improve the system throughput. Then, WMMSE algorithm is used to solve the non-convex power allocation problem. The local optimal solution can be obtained by low complexity iterative optimization. The convergence and effectiveness of the algorithm are verified by simulation. Aiming at the problem that the performance stability of NOMA beamforming system is affected by the CSI estimation error, the robust beamforming design in single-carrier NOMA with multiple input and single output channels is studied in this paper. Specifically, two types of CSI errors are considered: second order bounded CSI error and second order unbounded CSI error. In the bounded CSI error model with elliptic domain constraints, this paper presents the target rate maximization problem of user uninterrupted decoding under transmit power constraints. In order to solve this problem, it is first decomposed into a series of optimization problems with minimum power consumption. By using convex optimization theory, stable beamforming vector can be obtained, and high rate performance can be obtained with CSI error. In the second order unbounded CSI error model, this paper studies the optimization problem of minimizing transmission power with the quality of service satisfying the outage probability of the user. The constraint of outage probability makes the original problem difficult to deal with. By solving this problem, a robust beamforming vector which can satisfy the requirement of interrupt probability is obtained. Simulation results verify the robustness and effectiveness of the proposed algorithm.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【参考文献】