高能效的无线资源管理技术研究

发布时间：2018-07-03 20:26

本文选题：能量效率 + 资源分配　；参考：《东南大学》2016年博士论文

【摘要】：无线通信技术的发展和网络规模的扩大,导致了能量消耗爆炸式的增长,现有移动通信行业已经成为高能耗的产业。随着高速率无线通信业务需求的不断增长,通信系统的能耗会进一步增加。绿色节能的通信网络已经成为不可避免的发展趋势。另外,移动通信系统中的无线资源是有限的,并且日益紧缺,例如：功率资源、时间资源、频率资源、空间资源等。如何通过合理的资源配置获得更高的系统性能,一直是无线通信技术研究的重要内容。目前,高容量和高频谱效率不再是无线资源管理追求的单一目标。本论文主要从能量效率最优的角度,对无线资源管理问题进行了研究。通过对无线资源进行有效地配置,获得最优的系统能量效率,同时满足无线网络的各种需求。首先,研究了多天线多小区协作系统高能效的下行传输方法。在保证用户最小速率要求的前提下,以系统能量效率为目标,进行联合的波束成型和功率分配的设计。将问题描述为一个非线性的分数规划,并且包含非凸的约束条件。由于小区间干扰的存在和目标函数的分数结构,考虑采用多小区协作迫零的波束成型方案,消除干扰,从而简化目标函数和约束条件的非凸形式。采用参数化的转换对分数规划问题进行非分数的处理,并采用迭代算法计算辅助参数。考虑到协作迫零波束成型进行零空间的分解计算的复杂性,采用一种波束跟踪的方法进行简化处理。在给定波束成型矩阵后,求解了能效最优的功率分配问题。仿真结果表明,该方案的能量效率性能优于传统的最大化系统容量的传输方案,并且对于慢变信道,该方案能够以更低的复杂度获得与传统空间分解方法接近的能量效率性能。其次,研究了多小区正交频分多址接入(OFDMA, Orthogonal Frequency Division Multiple Access)系统高能效的资源分配方法。该方案基于非完美的信道状态信息进行资源分配的设计,并通过基站间的协作来处理同信道干扰。提出一种联合的用户调度、速率分配和功率分配算法,最大化系统能量效率。由于目标函数和约束条件的概率化限制,通过对约束条件的转化和近似,推导了速率变量与其他优化变量的函数关系。考虑到优化问题的混合非凸性质,即功率变量的连续性和用户调度变量的离散性,采用一种交替计算的迭代算法,并证明了算法的收敛性。然后分别计算了能效最优的用户调度和能效最优的功率分配。对功率分配子问题进行了下界缩放和参数化转换,并且当迭代过程收敛时下界是紧的。通过最优的能量效率的折衷函数曲线,分析了算法中最优能量效率的一些有用性质。仿真结果表明,虽然最优问题是非凸的,提出的方案可以得到与最优能效的上界接近的能效性能。与传统资源分配方案(如谱效最大化方案、不考虑信道误差的方案和等功率轮询调度方案等)相比,该方案可以获得更高的能效性能。同时,当存在同信道干扰时,提出的算法能够以很低的吞吐量损失,获得较大的能效增益。再次,研究了异构网络高能效的资源分配方法。该资源分配方案以权重能效为目标,即不同类型的基站具有不同的能效权重。将优化问题描述为一个非线性的分数求和优化问题,并且同时考虑每个用户的最小速率要求。为了处理优化问题的混合非凸形式,采用一种两步方案,即分开处理子信道分配和功率分配问题。初始化一个功率分配,提出一种启发式的子信道分配方案。首先找到距离速率要求最远的用户,再将该用户具有最高信干噪比的子信道分配给它。当所有用户的速率需求满足后,每个子信道上以最大的信干噪比的方式进行用户选择。对于给定的子信道分配,通过一阶近似将目标函数进行线性化,然后通过参数化方法对分数求和算法进行处理,并通过一个两层迭代算法获得功率分配。仿真结果表明,提出的算法需要很少的迭代次数就可以达到收敛,并且可以获得比全局能效最大化算法和谱效最大化算法更高的权重能效性能。另外,所提出的算法在宏基站和微微基站之间存在能量效率的折衷关系。然后,研究了中继辅助的OFDMA系统高能效资源分配问题。考虑多中继多用户的蜂窝系统,采用解码转发中继协议。在保证系统最小速率的前提下,该资源分配算法以能量效率最大化为目标。通过基站到中继和中继到用户两个时隙的联合速率分配、功率分配和子载波分配,获得最大的系统能量效率,保证最小速率和中断概率等服务质量要求。为了获得全局最优解,对优化问题的子载波分配变量进行了连续化松弛,使得所有优化变量(速率变量、功率变量、载波分配变量)统一为连续变量。可以证明,最终的载波分配可以获得和没有松弛时相同的性能,即保证方案的全局最优性。对优化问题进行了参数化的处理,并通过二分法进行了求解。由于在解码转发协议中,链路的容量为两时隙的链路容量的最小值,所以每次迭代中需要求解的问题是一个最大化最小问题。通过引入辅助变量将该问题转换为上镜图形式,并在对偶问题求解时通过最优性判断消除了新引入的辅助变量,获得全局最优的资源分配解。另外,采用了比例公平的机制进行了能效方案的设计,平衡系统的能量效率和用户的公平性。仿真结果表明,与传统方案相比,提出的方案可以获得更高的能效性能。同时,系统能效与谱效、系统能效与用户公平性、系统谱效与用户公平性存在一定的折衷关系。最后,研究了中继辅助的蜂窝系统低复杂度的高能效功率分配和速率自适应方案。考虑多中继多用户的蜂窝系统,采用前向放大中继协议。该资源分配以能量效率为目标,同时考虑系统的最小速率限制和最大发射功率限制等约束条件。通过两个时隙联合的功率和速率的分配设计,获得最大的系统能量效率和低复杂度最优的资源分配方案。考虑到问题的非凸性,利用高信噪比近似和参数化转换的方法,对优化问题进行了求解。一般地,分数规划问题需要迭代求解,并且在迭代过程中,需要多次求解多约束的优化问题。根据最优解的可行域范围,分别研究了该优化问题的最优解可能出现的几种形式,并分别进行了求解。然后利用二分法的上下界特殊的更新形式,以及最优能效与功率和速率的折衷关系,在每次迭代中判决最优解是否满足条件,从而进行判决循环跳出。该方案可以减少迭代次数,以及降低每次迭代优化问题的计算复杂度。仿真结果表明,该方案可以在不损失能效性能的前提下,降低算法的复杂度,同时保证获得全局最优的能量效率性能。尤其是当最大发射功率约束或最小速率约束要求严格时,提出的算法仅需要很少的迭代次数。另外,与谱效优先的算法和不考虑信道非完美性的方法相比,提出的方案可以获得比更高的能效性能。
[Abstract]:The development of wireless communication technology and the expansion of the network scale have led to the explosive growth of energy consumption. The existing mobile communications industry has become a high-energy consumption industry. With the increasing demand for high-speed wireless communication services, the energy consumption of communication systems will be further increased. The green and energy-saving communication network has become an inevitable issue. In addition, wireless resources in mobile communication systems are limited and are increasingly scarce, such as power resources, time resources, frequency resources, space resources, and so on. How to obtain higher system performance through reasonable allocation of resources has always been an important content of wireless communication technology research. At present, high capacity and high frequency spectrum efficiency are no longer available. This paper is the single goal of wireless resource management. This paper mainly studies the problem of wireless resource management from the perspective of the optimal energy efficiency. Through the effective configuration of wireless resources, the optimal system energy efficiency is obtained and the various needs of the wireless network are met. First, the multi antenna and multi cell collaboration system is studied. Under the premise of ensuring the minimum rate of users, the design of joint beamforming and power allocation is designed with the goal of system energy efficiency. The problem is described as a nonlinear fractional programming with a non convex constraint condition. Due to the existence of small interval interference and the fractional junction of the target function In order to eliminate interference and simplify the non convex form of the objective function and constraint condition, the method of parameterized conversion is used to deal with the fractional programming problem, and the auxiliary parameters are calculated by the iterative algorithm. A beam tracking method is used to simplify the processing. The power allocation problem with the best energy efficiency is solved after a given beamforming matrix. The simulation results show that the energy efficiency performance of the scheme is better than that of the traditional transmission scheme maximizing the system capacity. And for the slow changing channel, the scheme can be reduced to a lower complex. The heterozygosity obtains the energy efficiency performance that is close to the traditional spatial decomposition method. Secondly, the high efficiency resource allocation method of the OFDMA (Orthogonal Frequency Division Multiple Access) system is studied. The scheme is designed based on the imperfect channel state information, and through the base station. A joint user scheduling, rate allocation and power allocation algorithm is proposed to maximize the energy efficiency of the system. Due to the probability limitation of the target function and constraint conditions, the function relationship between the rate variable and the other optimal variables is derived by the transformation and approximation of the constraint conditions. The hybrid non convex property of the problem, that is, the continuity of power variables and the discreteness of the user scheduling variables, the convergence of the algorithm is proved by an iterative algorithm, and then the energy efficiency optimal user scheduling and energy efficiency optimal power allocation are calculated. The power gamete problem is reduced and parameterized. And when the iterative process converges, the lower bound is tight. Some useful properties of the optimal energy efficiency in the algorithm are analyzed by the optimal energy efficiency tradeoff function curve. The simulation results show that, although the optimal problem is non convex, the proposed scheme can get the energy efficiency close to the optimal energy efficiency. The scheme, such as the scheme of spectral efficiency maximization, the scheme of the channel error and the equal power polling scheduling, can obtain higher energy efficiency performance. At the same time, the proposed algorithm can gain more energy efficiency with low throughput loss when there is the same channel interference. Again, the high energy efficiency of the heterogeneous network is studied. Resource allocation method. The resource allocation scheme aims at weighting energy efficiency, that is, different types of base stations have different energy efficiency weights. The optimization problem is described as a nonlinear fractional sum optimization problem, and the minimum rate requirements for each user are considered at the same time. In order to deal with the mixed non convex form of the optimization problem, a kind of two is used. The step scheme is to separate the sub channel allocation and power allocation problem separately. Initialize a power allocation and propose a heuristic subchannel allocation scheme. First, the user with the farthest distance rate is found, and then the subchannel with the maximum signal to noise ratio is assigned to it. Each subchannel when the rate needs of all users are satisfied. The channel is selected in the way of the maximum signal to noise ratio. For given subchannel allocation, the target function is linearized by the first order approximation, then the fractional sum algorithm is processed by the parameterization method, and the power distribution is obtained by a two layer iterative algorithm. The simulation results show that the proposed algorithm needs little. The number of iterations can be convergent and the weight efficiency performance can be higher than the global energy efficiency maximization algorithm and the spectral efficiency maximization algorithm. In addition, the proposed algorithm has the tradeoff between the energy efficiency between the Acer station and the micro base station. Then, the high efficiency resource allocation problem of the secondary assisted OFDMA system is studied. Based on the minimum rate of the system, the resource allocation algorithm aims at maximizing the energy efficiency. The maximum system energy is obtained from the base station, the relay and the relay to the user's two time slots, the power allocation and the subcarrier allocation. In order to obtain the global optimal solution, in order to obtain the global optimal solution, the subcarrier allocation variable of the optimization problem is continuous relaxation, so that all the optimization variables (rate variables, power variables, carrier allocation variables) are unified as continuous variables. It can be proved that the final carrier allocation can be obtained and failed. The same performance, that is, to ensure the global optimality of the scheme, is to ensure the global optimality of the scheme. The optimization problem is parameterized and solved by a dichotomy. Since the capacity of the link is minimum of the link capacity of the two slot in the decoding and forwarding protocol, the problem that needs to be solved in each iteration is a maximum minimum problem. By introducing the auxiliary variable, the problem is converted to the form of the upper mirror, and the newly introduced auxiliary variable is eliminated by the optimality judgment in the solution of the dual problem, and the global optimal resource allocation solution is obtained. In addition, the energy efficiency scheme is designed to balance the energy efficiency of the system and the fairness of the user by using the mechanism of proportional fairness. The simulation results show that compared with the traditional scheme, the proposed scheme can obtain higher energy efficiency. At the same time, there is a tradeoff between system energy efficiency and spectral efficiency, system energy efficiency and user fairness, system spectral efficiency and user fairness. Finally, the high energy efficient power allocation and rate self distribution of the low complexity of the relay assisted bee nest system are studied. Adaptive scheme. Considering the multi relay and multiuser cellular system, the forward amplification relay protocol is adopted. The resource allocation is aimed at energy efficiency, and the minimum rate constraints and maximum transmission power constraints are considered. The maximum system energy efficiency is obtained through the allocation of the power and rate of two slots combined with the power and rate. In consideration of the non convexity of the problem, the optimization problem is solved by using the method of high signal-to-noise ratio approximation and parameterized conversion. Generally, the fractional programming problem needs to be solved iteratively. In the iterative process, the optimization problem of multiple constraints needs to be solved many times. According to the feasible domain model of the optimal solution. Several forms of the optimal solution of the optimization problem are studied and solved respectively. Then, the special updating form of the upper and lower bounds of the dichotomy, and the tradeoff between the optimal energy efficiency and the power and rate are used, and the decision of the optimal solution satisfies the condition in each iteration, thus the decision cycle is jumped out. It can reduce the number of iterations and reduce the computational complexity of each iterative optimization problem. The simulation results show that the proposed scheme can reduce the complexity of the algorithm without loss of energy efficiency and guarantee the energy efficiency performance of the global optimal. Especially when the maximum emission power constraint or minimum rate constraint is strict. The proposed algorithm requires only a small number of iterations. In addition, the proposed scheme can obtain higher energy efficiency compared with the spectral efficiency priority algorithm and the method that does not consider the channel imperfections.
【学位授予单位】：东南大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN92

【相似文献】