微波通信链路中速率自适应算法研究

发布时间：2018-02-03 19:35

本文关键词： 速率自适应调制编码方式多输入多输出聚合的MAC层协议数据单元　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：由于微波通信中无线链路不可靠,节点在发送数据时所选的发送速率和传输模式并不一定适应当前信道传输状况。当无线链路质量良好时,如果采用低的速率发送数据,将会导致网络吞吐量降低;然而当无线链路质量较差时,如果采用高的数据发送速率,将会导致高的误比特率(BER)。因而节点需要在传输过程中实时计算并预测信道质量变化,采取适应当前信道的数据传输速率,即速率自适应。速率自适应算法应用于媒质接入控制(MAC)层,在数据传输前决定其发送速率以及传输模式。数据发送速率由物理层提供的不同等级的调制编码策略(MCS)选定,而传输模式则由物理层的多输入多输出(MIMO)模式实现:空间分集(SD)或者空间复用(SM)。速率自适应对提高无线资源利用率和整个网络的性能有着重要的影响。根据是否依赖反馈信息,速率自适应可以分为开环的速率自适应算法和闭环的速率自适应算法。每类算法各有其特点,开环速率自适应算法无需更改协议,实现简单;而闭环速率自适应算法对信道的预估相对更加准确。即使同一类中,不同的速率自适应算法,其性能也各有差异,没有任何一种算法能完美适应所有应用场景。本文提出了用于IEEE802.11n标准下兼容AMPDU的自适应多速率和变聚合度重传算法AMRAL。算法主要包括信道质量预测机制、速率调整机制、聚合度调整机制和A-RTS/CTS机制四部分,通过分析Bitmap,获取丢失分组在AMPDU中分布情况来估计信道质量的变化趋势,同时根据对信道质量好坏和变化趋势的预测,将数据传输速率和聚合帧长度结合进行调整。AMRAL算法的创新点在于首次提出了对聚合帧的长度进行自适应调整以适应信道传输状况,同时还在于通过分析Bitmap来预测信道变化趋势,从而使得算法对信道状况的预估更加合理准确。通过对数据传输速率、聚合帧长度、MIMO模式的联合自适应调制,AMRAL算法能更加有效地利用信道。通过在网络仿真软件NS-3中实现AMRAL、Atheros和MiRA算法,本文对这三种速率自适应算法的性能进行了对比分析。其中Atheros和MiRA算法已经在硬件驱动中实现并且广泛应用。在算法收敛性方面,仿真结果表明AMRAL算法比Atheros和MiRA算法能够更快地收敛到适合当前信道的传输速率;在信道适应性方面,AMRAL算法表现良好,特别是在长距离通信场景中;而在网络吞吐量方面,仿真结果显示AMRAL可以有效提升传输空口时间内信道利用率。
[Abstract]:Due to the unreliable wireless link in microwave communication, the transmission rate and transmission mode selected by the node when transmitting data are not necessarily adapted to the current channel transmission condition. When the wireless link quality is good. If the data is sent at a low rate, the network throughput will be reduced. However, when the wireless link quality is poor, if the high data transmission rate is adopted, it will lead to a high bit error rate (BER). Therefore, the node needs to calculate and predict the channel quality changes in real time during the transmission process. The data transmission rate adapted to the current channel, that is, rate adaptive. Rate adaptive algorithm, is applied to the medium access control (MAC) layer. The transmission rate and the transmission mode are determined before the data transmission.; the data transmission rate is selected by the different levels of modulation and coding policies provided by the physical layer (MCSs). The transport mode is implemented by the multi-input and multi-output MIMO-mode of the physical layer: spatial diversity (SDN) or spatial multiplexing (SMM)). Rate adaptation plays an important role in improving the utilization of wireless resources and the performance of the network as a whole, depending on whether or not it depends on feedback information. Rate adaptation can be divided into open-loop rate adaptive algorithm and closed-loop rate adaptive algorithm. The closed-loop rate adaptive algorithm is more accurate to predict the channel. Even in the same class, the performance of different rate adaptive algorithm is different. No one algorithm is perfect for all application scenarios. This paper presents an adaptive multi-rate and variable degree of aggregation retransmission algorithm AMRAL for AMPDU compatible with IEEE802.11n standard. The algorithm mainly includes channel quality prediction mechanism. Rate adjustment mechanism, degree of polymerization adjustment mechanism and A-RTS / CTS mechanism, through the analysis of Bitmap. The distribution of the lost packets in the AMPDU is obtained to estimate the variation trend of channel quality, and to predict the channel quality and the changing trend according to the channel quality. The innovation of combining the data transmission rate with the aggregate frame length to adjust .AMral algorithm is that the adaptive adjustment of the aggregate frame length is proposed for the first time to adapt to the channel transmission situation. At the same time, the channel change trend is predicted by analyzing Bitmap, which makes the estimation of the channel condition more reasonable and accurate. Through the data transmission rate, aggregate frame length. The joint adaptive modulation algorithm of MIMO mode can make use of the channel more effectively. The Atheros and MiRA algorithms are implemented in the network simulation software NS-3. In this paper, the performance of these three rate adaptive algorithms is compared and analyzed. Among them, Atheros and MiRA algorithms have been implemented in the hardware driver and widely used. The simulation results show that the AMRAL algorithm can converge faster than the Atheros and MiRA algorithms to the transmission rate suitable for the current channel. In the aspect of channel adaptability, the AMral algorithm performs well, especially in long distance communication scenarios. In terms of network throughput, the simulation results show that AMRAL can effectively improve the channel utilization in transmission space time.
【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】：TN925

【相似文献】