卡尔曼滤波器在带宽测量中的应用

发布时间：2018-04-26 07:32

本文选题：带宽测量 + 卡尔曼滤波器　；参考：《电子科技大学》2015年硕士论文

【摘要】：带宽测量在网络信息领域中是一项非常重要的技术。它通过测量出网络路径在单位时间内能传输的最大数据量,来指导网络系统的Qo S管理,拥塞控制,以及路由选择访问等。2006年,Ekelin S,Nilsson M等研究学者提出了在带宽测量领域中使用卡尔曼滤波器的测量方法BART,在保持高测量精度的同时,提高了对变化环境的跟踪性能,不仅扩大了算法的使用条件,也大幅度降低了计算功耗。2008年,采用了BART设计思路的基于概率模型的ABEST算法也被提出,拓展了卡尔曼滤波器在带宽测量领域的使用。不过,在BART和ABEST中均只提出了卡尔曼滤波器的使用方法,当链路环境改变时,若使用相同的参数设置则可能导致测量的结果会千差万别。本文在研究卡尔曼滤波器的过程中,先以BART算法为基础,通过模型,理论,实验分析,引入了卡尔曼滤波器的自适应参数设定,提高了算法对不同链路环境的适应性,并提高测量的精度和跟踪性能。接着以研究ABEST算法为基础,根据标准状态方程的完整性,提出了动态状态方程的改进思路,设计了双探测流发包策略,提高了算法的跟踪性能。为了补充卡尔曼滤波器在带宽测量领域中的应用,以及考虑结合两种改进算法,本文在基于探测间隔模型IGI中,建立卡尔曼滤波器以脱离对链路容量C的依赖,并尝试将自适应参数设定和动态状态方程结合起来,进一步提高卡尔曼滤波器系统的测量精度,跟踪性能,稳定性,以及实用性。
[Abstract]:Bandwidth measurement is a very important technology in the field of network information. It can direct QoS management and congestion control of network system by measuring the maximum amount of data that network path can transmit in unit time. In 2006, Ekelin Schion Nilsson M and other researchers put forward a measurement method, Bart, which uses Kalman filter in bandwidth measurement field, which not only keeps high measurement accuracy, but also improves the tracking performance of changing environment. In 2008, a probabilistic model based ABEST algorithm based on BART was proposed, which extends the use of Kalman filter in bandwidth measurement. However, in both BART and ABEST, only the Kalman filter is used. When the link environment changes, if the same parameters are used, the measurement results may vary greatly. In this paper, based on BART algorithm, the adaptive parameter setting of Kalman filter is introduced through model, theory and experiment analysis, which improves the adaptability of the algorithm to different link environment. And improve the measurement accuracy and tracking performance. Then, based on the research of ABEST algorithm, according to the integrity of the standard equation of state, the improved idea of dynamic state equation is put forward, and a dual-probe packet sending strategy is designed to improve the tracking performance of the algorithm. In order to supplement the application of Kalman filter in bandwidth measurement and to consider two improved algorithms, in this paper, based on the detection interval model (IGI), a Kalman filter is established to get rid of the dependence on link capacity C. The adaptive parameter setting and dynamic state equation are combined to improve the measurement accuracy, tracking performance, stability and practicability of Kalman filter system.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN713

【参考文献】