【作者】 沈敏洁；

【导师】 赵晓晖；

【作者基本信息】 吉林大学 ， 信号与信息处理， 2009， 硕士

【摘要】 随着移动通信和互联网的高速发展,无线多媒体业务的需求不断增长,人们对宽带无线接入的需求越来越迫切。故下一代移动通信系统应用了多种先进的传输技术,以解决人们对带宽的需求和有限的无线资源之间的矛盾。其中物理层中的OFDM技术和MIMO技术受到广泛关注。OFDM可通过将频率选择性多径衰落信道在频域内转变为平坦信道,减小了多径衰落的影响, MIMO技术能够在空间中产生同时传输多路数据流的独立并行信道,直接有效地增加了系统的传输效率,成倍地提高信道容量。正因为MIMO-OFDM技术能带来系统性能的提高,它被公认为新一代无线通信网络主流的信息传输处理技术,这使得无线资源管理面临许多新的问题。无线资源管理的目标是通过在有限带宽的条件下对无线通信系统中有限的无线资源进行协调和管理,控制整个系统的状态,最大化系统资源的利用效率,为网络内无线用户终端提供业务质量保障。本文在从物理层和MAC层相结合的思路出发,提出了一种基于空、时、频三维分配的自适应OFDMA下行链路资源调度和分配算法。该算法不仅结合了先进的物理层技术,同时从MAC层考虑了不同业务用户物理信道状态信息以及用户速率要求、延时要求、等QoS参数,并且兼顾了系统频谱利用效率。在物理层通过天线选择算法和自适应调制技术对系统进行载波和比特分配,出于系统性能增益、算法计算复杂度以及控制开销方面的考虑,物理层功率分配算法采用了等功率分配方法,并在接收端通过反馈信道将所得到的信道状态信息告知基站端的MAC层包调度器和物理层的资源分配模块,调度器根据CSI和相关QoS信息合理地对缓存队列里的包和用户进行优先级的排列和调度,在保证满足用户QoS的前提下提高信道利用率,最终使得系统吞吐量最大化。仿真结果表明,该算法随着数据用户的增多,能够有效地支持数据用户业务的传输,在保证多业务传输质量的同时提高系统吞吐量,实现用户间的相对公平性。更多还原

【Abstract】 With the rapid development of wireless communication and Internet, The modes of new generation mobile communication networks become diverse, high-speed, flexible, and high efficient. The target of broad band wireless communication is to transmit data fast and reliably, while there are two tough problems in front of us, they are multipath fading channel and bandwidth efficiency. People now is looking forward to the higher data transmission speed and higher system capacity offered by the next generation mobile communication system in order to meeting the different QoS requests in different applications with more users. Therefore, the next generation mobile communication system adopts many advanced transmission techniques to abate the contradiction between the requirements of bandwidth and limited wireless resources. Thus Orthogonal Frequency Division Multiplexing (OFDM) technique and Multiple Input Multiple Output (MIMO) technique has been getting more attention.OFDM technique changes the frequency selective multipath fading channel into flat fading channel in frequency domain which effectively mitigates the effects of multipath propagation, hence, it increases data rate. In additional, MIMO technique can independently generate parallel multi-channel data stream in spatial at the same time, which directly and effectively increases transmission efficiency. Therefore, combination of MIMO and OFDM technique is believed to have the ability to achieve high transmission speed and strong reliability.Since OFDM and MIMO technique can improve system performance, they have been acknowledged as two major information transmission and processing techniques for next generation wireless communication networks. Thus there are many problems facing to radio resource management. The target of radio resource management includes control of the state of whole system, maximum of the utilized ratio of the system resources, and guarantee of the QoS requirement for user terminals in wireless networks by modulating and managing the limited radio resource with limited bandwidth. Due to the limited radio resource and the QoS requirement (as delay and throughput) of multimedia applications for the downlink of MIMO-OFDMA system, how we can support different applications (Voice, Video or Data, etc), optimize the system resource, and offer the QoS guarantee for different applications in multi-user system are becoming the focuses in both It industry and academic study presently with the development of telecommunication technology.The scheduling and resource allocation algorithm in MIMO-OFDMA system is divided into three categories, (1) Fairness and QoS guarantee based approach, such as Round-Robin algorithm, can provide equal service opportunity for users by sacrificing the throughput of system. And Max C/I Algorithm can achieve the best system throughput, but the users whose channel condition is better always hold the priority, and result in unfair scheduling between users. Then PF Algorithm guarantees the users’fairness, but the delay is too long to tolerate by real time users. (2) MIMO/OFDM space-time-frequency domain resource management is the method that, according to the QoS requirement, it is based on the modeling for OFDMA system subcarriers, bit or power, then builds a object function with some restrictions, finds a method to resolve the equation. But its INI (interfere) and QoS requirement make the calculation much more complex, and the mathimatical model is usually nonlinear combination optimization problem. (3) Joint optimization plan in which we can make a subchannel on frequency domain multiply a OFDM symbol on time domain as a least resource assignment unit, at the same time we can consider PHY layer channel information and MAC layer QoS requirements in the same time. Then we can set respective priority according to users’different data stream types with not only the design of PHY layer but also the calculation of different applications which can better benefit from layers optimization. Therefore cross-layer planning can achieve better system performance through respective function of different layer.2. Research on A Scheduling and Resource Allocation Algorithm for Downlink of MIMO-OFDMA SystemIn this paper, a novel QoS-guarantee efficient scheduling and resource allocation algorithm for heterogeneous traffics in the downlink of MIMO-OFDMA system is proposed to overcome the disadvantage of original algorithms, which can sufficiently exploit space,time, frequency resource to provide QoS-guarantee. The algorithm considers not only the advanced physical techniques but also the service characteristics, QoS requirements, and fairness for users observed in MAC layer.The basic idea is using time-frequency block as an allocation unit to decrease calculation time, then employing TAS (Transmit Antennas Selection) and AM (Adaptive Modulate) technique in PHY layer to allot subcarriers and bits. Taking whole cost, system performance gain and complexity of algorithm into consideration, transmit power is equally allotted for each subcarriers. Scheduler and resource allocater can get the CSI from receiver through feedback channel. Scheduler schedules packets queue in buffer and updates priority of different users to improve utilization rate of channel and finally maximize system throughput.3. Simulation results and AnalysisThe simulations of comparative experiments of the improved algorithm on computer with MATLAB are performed. Firstly we present the modeling plan and a multiuser MIMO-OFDM system is considered here. The proposed improved method is used in SISO and MIMO system and compared with the average throughput of system. Then the original method in reference [51] and the optimal method for heterogeneous traffics in MIMO-OFDMA system are compared in different aspects, such as throughput average delay of VoIP and users fairness.Simulation results show that the proposed method and the original method can achieve same system throughput performance when the system has few users, but when the number of users increases rapidly, the proposed method can significantly improve the system throughput performance while guarantees efficient transmission for heterogeneous traffics. We must admit that the original method can get better performance in average delay of VoIP since the proposed algorithm gets the advantage of system throughput by sacrificing the delay of real time users. However we can still guarantee the delay (about 10ms) that is less than the Delay-bound of VoIP.更多还原