基于UWB的室内定位算法研究与应用

发布时间：2018-01-21 23:13

本文关键词： UWB 无线传感器网络卡尔曼滤波 C# 混合编程　出处：《山东大学》2014年硕士论文　论文类型：学位论文

【摘要】：近年来,随着无线传感器网络的飞速发展,无线定位技术也越来越受到人们的关注。传统的定位如红外线、蓝牙、超声波、RFID等技术,在定位精度、系统功耗等方面不能满足人们的需求。而超宽带(Ultra Wide Band, UWB)技术具有传输速率高、发射功率低、抗干扰和穿透能力强等特点,特别是UWB定位具有非常高的定位精度,因此对UWB定位技术的研究具有重要的应用价值。本文通过对UWB定位平台的需求分析和实际硬件情况,分别进行了算法和上位机软件的研究应用,并对系统架构进行了设计。在定位算法方面,本文在UWB定位系统的平台上,对Fang算法、Chan算法等经典定位算法以及分别基于TOA和TDOA的扩展卡尔曼滤波算法进行了仿真研究,针对传播过程中环境等因素造成的非视距误差,运用有偏卡尔曼滤波算法进行了非视距误差的抑制。仿真发现传统定位算法在可视距距差较小时能达到比较高的精度,但是随着误差的增大,算法性能有着显著下降。其中,Fang算法不能利用冗余信息提高定位精度,Chan算法的性能在非视距环境中显著下降,而卡尔曼滤波算法能有效地抑制测量误差和非视距误差的影响。本文重点研究了卡尔曼滤波的定位算法和非视距误差的抑制算法,首先对初始到达时间测量值进行有偏卡尔曼滤波抑制其非视距误差,然后利用处理后的测量值进行扩展卡尔曼滤波定位。后续章节的仿真和应用实例表明,该算法比传统定位算有非常好的定位效果,在非视距误差环境中也能达到较高的定位精度。本文的主要目的在于将卡尔曼滤波算法应用于PLUS定位平台上,基于Visual Studio2012开发环境,使用C#和MATLAB混合编程的方式实现,由于MATLAB提供了相当丰富的矩阵运算函数,对卡尔曼滤波算法非常实用,卡尔曼滤波算法经由MATLAB编译为动态链接库文件,执行效率高,易于维护。软件采用C#编程来达到对移动台的实时定位,软件运行过程中实现了数据处理、计算机绘图、数据保存、数据显示以及多项任务的实时并行处理,提高了数据处理的效率。实验结果表明,本文设计的定位系统能满足定位的需要,软件运行稳定且实时性好,所设计的算法提高了定位精度,降低了UWB系统的定位误差。
[Abstract]:In recent years, with the rapid development of wireless sensor networks, wireless positioning technology has been paid more and more attention. Traditional positioning technology, such as infrared, Bluetooth, ultrasonic RFID and other technologies, in the positioning accuracy. The system power consumption and other aspects can not meet the needs of people, while ultra-wideband (UWB) technology has high transmission rate and low transmission power. The characteristics of anti-jamming and strong penetration, especially UWB positioning has a very high positioning accuracy, so the study of UWB positioning technology has an important application value. In this paper, the requirements of the UWB positioning platform and the actual hardware situation, respectively, algorithms and host computer software research and application, and the design of the system architecture. In the positioning algorithm. In this paper, on the platform of UWB positioning system, the classical localization algorithms such as Fang algorithm and extended Kalman filter algorithm based on TOA and TDOA are simulated. Aiming at the non-line-of-sight error caused by environmental factors in the process of propagation, the offset Kalman filter algorithm is used to suppress the non-line-of-sight error. Simulation results show that the traditional positioning algorithm can achieve a higher accuracy when the visual distance difference is small, but with the increase of the error, the performance of the algorithm has a significant decline. The Fang algorithm can not use redundant information to improve the positioning accuracy. The performance of the Fang algorithm is significantly reduced in the non-line-of-sight environment. The Kalman filter algorithm can effectively suppress the influence of measurement error and non-line-of-sight error. This paper focuses on the Kalman filter location algorithm and non-line-of-sight error suppression algorithm. First, the non-line-of-sight error is suppressed by biased Kalman filter, and then the extended Kalman filter is used to locate the measured value. The simulation and application examples of the following chapters show that the proposed method can be used to solve the problem of non-line-of-sight error. The algorithm has a better localization effect than the traditional positioning algorithm, and it can also achieve high positioning accuracy in the non-line-of-sight error environment. The main purpose of this paper is to apply the Kalman filter algorithm to the PLUS location platform, based on the Visual Studio2012 development environment. Using C # and MATLAB mixed programming, because MATLAB provides a lot of matrix operation functions, it is very practical to Kalman filter algorithm. The Kalman filter algorithm is compiled into dynamic link library file by MATLAB, which is efficient and easy to maintain. The software uses C # programming to locate the mobile station in real time. Data processing, computer drawing, data saving, data display and real-time parallel processing of many tasks are realized in the process of software operation, and the efficiency of data processing is improved. The experimental results show that the positioning system designed in this paper can meet the needs of location, and the software runs stably and real-time. The algorithm designed improves the positioning accuracy and reduces the positioning error of UWB system.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN925

【参考文献】