多小区OFDMA系统资源分配算法与信道估计

发布时间：2018-04-23 09:35

本文选题：多小区OFDMA + 资源分配　；参考：《华南理工大学》2014年博士论文

【摘要】：随着无线通信应用的丰富和发展，服务质量(QoS)需求与无线通信资源有限的矛盾日益尖锐。为缓解这一矛盾，必须从无线通信系统的各个层面展开研究。在物理层，正交频分复用(OFDM)是一种非常适合于宽带无线系统的技术;在链路层，采用资源分配和调度可以显著改善通信系统的频谱效率和吞吐量。另外，OFDMA技术的独特优势，为资源分配提供了灵活的自由度，因此，OFDMA系统中的资源分配算法得到了广泛的关注。正交频分多址(OFDMA)是将正交频分复用(OFDM)技术和频分多址(FDMA)技术相结合的多址方案，是第四代移动通信系统(4G)的关键技术之一。OFDMA以OFDM调制为基础，通过给不同用户独立地分配子载波来实现多用户接入，在多用户同时接入的情况下，OFDMA系统基于无线信道的状态、服务质量(QoS)要求等参数动态地制定资源配置方案，从而实现资源的自适应最优配置。这样，系统既能获得更高的频谱利用率，又可以更好地满足QoS要求。自适应资源配置是当前一个重要的研究课题，，也是OFDMA系统急需解决的核心问题之一。针对这个研究方向，本文在深入研究多小区OFDMA自适应资源配置过程中提出了一种基于罚函数的资源分配算法和一种低复杂度资源分配算法，同时研究了系统中信道噪声的估计，提出了一种基于最大似然法的联合测量干扰和噪声功率水平的非数据辅助算法。全文系统地分析和研究了在多小区OFDMA系统进行自适应资源配置需要解决的问题，具体主要有如下三个方面的工作和创新: (1)基于罚函数的模拟退火资源分配算法，在多小区OFDMA系统中，基于集中式资源管理，考虑本小区对其他小区的干扰情况，对各个小区的子载波和功率分配进行调整，从而达到最小化总功率的目的。通过功率离散化来建立多小区OFDMA问题模型，使用罚函数法来简化问题模型，并采用改进的模拟退火算法进行求解。本文提出的罚函数SA（Penalty-Simulation Annealing）资源分配算法，简化了问题模型，从而降低了模型求解复杂度。理论分析和仿真结果表明，离散功率个数的选择在一定范围内具有随机性。文章提出的算法获得了更好的系统吞吐量，同时有效降低了求解复杂度，且对整体性能不会产生不良影响；与多分配算法相比，单位功率吞吐量有显著提高。 (2)提出了一种低复杂度资源分配算法，通过对一组线性方程组的求解，在最小化传输功率的同时满足了用户最大传输需求。该算法把子载波和功率分配分别实施，即把资源分配过程分成子载波分配和功率分配两个独立的步骤进行，在子载波分配时，假设频效率固定简化了分配复杂度，在固定频率效率的情况下，功率分配可以通过一组线性方程组求得，当方程组的解出现负值时，通过调整算法对负值的解进行处理实现正确的分配结果，仿真结果表明该算法在系统吞吐量、功率效率、算法复杂度等方面，取得了一个介于最优算法和多分配算法之间的折衷。 (3)提出了一个基于最大似然法的平坦信道估计算法，该算法是一种联合测量干扰和噪声功率水平的非数据辅助算法。针对确定性未知高斯白噪声干扰的有限状态离散信号，该算法首先假设在特定条件下，得到一个近似的闭形式解，然后以些解为初始值，采用迭代的方法，以最大似然准则为基础，得到完善的估计结果。仿真结果表明该方法具有收敛速度快，准确性高等特性。
[Abstract]:With the enrichment and development of wireless communication applications, the contradiction between the quality of service (QoS) demand and the limited wireless communication resources is becoming increasingly acute. In order to alleviate this contradiction, it is necessary to study from the various levels of the wireless communication system. In the physical layer, orthogonal frequency division multiplexing (OFDM) is a technology that is not often suitable for broadband wireless systems; in the link layer, in the link layer, Resource allocation and scheduling can significantly improve the spectrum efficiency and throughput of the communication system. In addition, the unique advantage of OFDMA technology provides a flexible degree of freedom for resource allocation. Therefore, the resource allocation algorithm in the OFDMA system has been widely concerned.
Orthogonal frequency division multiple access (OFDMA) is a multiple access scheme combining orthogonal frequency division multiplexing (OFDM) technology and frequency division multiple access (FDMA) technology. It is one of the key technologies of the fourth generation mobile communication system (4G)..OFDMA is based on OFDM modulation. By assigning subcarriers to different users independently, multi user access is realized and the situation of simultaneous access to multi users is achieved. The OFDMA system is based on the state of the wireless channel, the quality of service (QoS) and other parameters to dynamically formulate the resource allocation scheme, thus realizing the optimal allocation of resources. In this way, the system can obtain higher spectrum utilization and better meet the requirements of QoS. Adaptive resource allocation is an important research topic at present. It is also one of the core problems needed to be solved urgently in the OFDMA system.
Aiming at this research direction, a resource allocation algorithm based on penalty function and a low complexity resource allocation algorithm are proposed in this paper. At the same time, the estimation of channel noise in the system is studied. A joint measurement of interference and noise power based on maximum likelihood method is proposed. The paper systematically analyzes and studies the problems that need to be solved in the self-adaptive resource allocation in multi cell OFDMA system, including the following three aspects: work and innovation.
(1) based on the penalty function based simulated annealing resource allocation algorithm, in the multi cell OFDMA system, based on centralized resource management, considering the interference of the community to other communities, the subcarrier and power allocation of each cell is adjusted to minimize the total power, and the multi cell OFDMA problem is established by power discretization. The penalty function method is used to simplify the problem model, and the improved simulated annealing algorithm is used to solve the problem. The penalty function SA (Penalty-Simulation Annealing) resource allocation algorithm is proposed in this paper, which simplifies the problem model and reduces the complexity of the model. The theoretical analysis and simulation results show that the selection of the discrete power number is in the first place. The proposed algorithm achieves better system throughput, reduces the complexity effectively, and does not have a bad effect on the overall performance. Compared with the multi allocation algorithm, the unit power throughput is significantly improved.
(2) a low complexity resource allocation algorithm is proposed. By solving a set of linear equations, the maximum transmission demand is met while the transmission power is minimized. The algorithm carries out the sub carrier and power allocation separately, which is divided into two independent steps, which are subcarrier allocation and power allocation. When the carrier distribution is distributed, it is assumed that the frequency efficiency is fixed and the allocation complexity is simplified. In the case of fixed frequency efficiency, the power allocation can be obtained by a set of linear equations. When the solution of the equations appears negative, the correct allocation results are processed by the adjustment algorithm to deal with the negative value. The simulation results show that the algorithm is used in the system throughput. In terms of quantity, power efficiency and algorithm complexity, a trade-off between optimal algorithm and multi allocation algorithm is obtained.
(3) a flat channel estimation algorithm based on maximum likelihood method is proposed. This algorithm is a non data auxiliary algorithm combined to measure the level of interference and noise power. According to the finite state discrete signal with uncertain Gauss white noise interference, the algorithm first assumes an approximate closed form solution under the definite condition. This method is based on the maximum likelihood criterion by using some solutions as the initial value. The simulation results show that the method has fast convergence speed and high accuracy.

【学位授予单位】：华南理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN929.5

【参考文献】