基于信息柔性融合的室内定位系统研究与实现
发布时间：2019-03-10 21:35
【摘要】：随着移动互联网行业的高速发展,人们对于室内定位的需求日益迫切,作为位置信息服务最重要的组成部分,室内定位技术逐渐受到越来越多的商业关注。由于技术的进步,人们对于室内定位的精度要求也越来越高,但是由于室内环境的反射和多径传播,高精度室内定位问题一直是定位领域的最大难题。论文以信息柔性融合技术为基础,主要研究如何在复杂室内环境下得到良好的室内定位结果。论文所做的主要工作如下:(1)论文对室内定位技术的研究现状进行了总结,剖析了各种室内定位技术的优缺点和技术难点。基于克拉美罗界,分析了TOA定位算法的理论极限,并推导了基站噪声特性不一致时TOA算法的CRLB。针对制约室内定位精度提高的瓶颈问题,设计了基于信息柔性融合的室内定位系统框架与结构。(2)针对信息柔性融合的数据层融合算法,分别研究了单/多传感器条件下的定位参数估计、多定位参数融合定位等问题。在此基础上提出了一种基于RSSI/AOA的新型室内精确定位方法,仿真和测试结果显示本文所给算法可以有效滤除参数测量波动造成定位结果误差,有效提高了室内复杂环境下的目标定位精度。(3)建立了基于强跟踪Kalman滤波器的决策层融合算法。解决Kalman滤波器在运动目标定位跟踪方面的缺陷:对突变状态的跟踪能力差,对于大噪声的滤除能力弱。改进强跟踪算法中的次优渐消因子,根据前次结果对强跟踪结果做反馈,形成了基于指数渐消因子的融合跟踪算法(EFF-STF)。详细说明了算法的实现步骤,建立系统实际采集数据验证了算法的有效性,并和相同条件下和卡尔曼滤波性能对比验证了算法在大噪声滤除方面的优越性,通过跟踪前后的定位误差CDF图验证了跟踪性能对定位误差的良好修正作用。(4)论文利用MATLAB和VS2010联合编程,实现了基于低功耗蓝牙的室内定位系统,将定位结果可视化,验证了算法的性能,实验测试表明经过信息柔性融合和跟踪处理之后,可以将85%的定位误差约束在1m以内,实现了室内实时精确定位的需求。
[Abstract]:With the rapid development of mobile Internet industry, the demand for indoor positioning becomes more and more urgent. As the most important part of location information service, indoor positioning technology has attracted more and more commercial attention. Due to the progress of technology, the accuracy of indoor positioning is more and more demanding. However, due to the reflection and multipath propagation of indoor environment, the problem of high-precision indoor positioning has always been the biggest problem in the field of positioning. Based on information flexible fusion technology, this paper mainly studies how to obtain good indoor positioning results in complex indoor environment. The main work of this paper is as follows: (1) the research status of indoor positioning technology is summarized, and the advantages and disadvantages and technical difficulties of various indoor positioning techniques are analyzed. Based on the Caramello bound, the theoretical limit of the TOA localization algorithm is analyzed, and the CRLB. of the TOA algorithm is derived when the noise characteristics of the base station are inconsistent. In order to solve the bottleneck problem which restricts the improvement of indoor positioning accuracy, the frame and structure of indoor positioning system based on information flexible fusion are designed. (2) the data layer fusion algorithm for information flexible fusion is proposed. The estimation of location parameters and fusion localization of multi-location parameters under single / multi-sensor conditions are studied respectively. On this basis, a new precise indoor location method based on RSSI/AOA is proposed. The simulation and test results show that the proposed algorithm can effectively filter the error caused by parameter measurement fluctuation. The accuracy of target location in complex indoor environment is improved effectively. (3) A decision-level fusion algorithm based on strong tracking Kalman filter is proposed. The defects of Kalman filter in moving target location and tracking are solved: poor tracking ability for abrupt state and weak filtering ability for large noise. A fusion tracking algorithm based on exponential fading factor (EFF-STF) is developed by improving the suboptimal fading factor in the strong tracking algorithm and feedback the strong tracking results according to the previous results. The implementation steps of the algorithm are described in detail, and the validity of the algorithm is verified by the establishment of the system's actual data collection, and the superiority of the algorithm in large noise filtering is verified by comparing the performance of the algorithm with the Kalman filter under the same conditions. The positioning error CDF diagram before and after tracking verifies the good correction effect of tracking performance on the positioning error. (4) in this paper, the indoor positioning system based on low power Bluetooth is realized by using MATLAB and VS2010 joint programming, and the positioning results are visualized. The performance of the algorithm is verified. The experimental results show that after information flexible fusion and tracking processing, 85% of the positioning error can be limited to less than 1m, and the requirement of real-time and accurate indoor positioning can be realized.
【学位授予单位】：河南师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN713;TN925
本文编号：2438052
[Abstract]:With the rapid development of mobile Internet industry, the demand for indoor positioning becomes more and more urgent. As the most important part of location information service, indoor positioning technology has attracted more and more commercial attention. Due to the progress of technology, the accuracy of indoor positioning is more and more demanding. However, due to the reflection and multipath propagation of indoor environment, the problem of high-precision indoor positioning has always been the biggest problem in the field of positioning. Based on information flexible fusion technology, this paper mainly studies how to obtain good indoor positioning results in complex indoor environment. The main work of this paper is as follows: (1) the research status of indoor positioning technology is summarized, and the advantages and disadvantages and technical difficulties of various indoor positioning techniques are analyzed. Based on the Caramello bound, the theoretical limit of the TOA localization algorithm is analyzed, and the CRLB. of the TOA algorithm is derived when the noise characteristics of the base station are inconsistent. In order to solve the bottleneck problem which restricts the improvement of indoor positioning accuracy, the frame and structure of indoor positioning system based on information flexible fusion are designed. (2) the data layer fusion algorithm for information flexible fusion is proposed. The estimation of location parameters and fusion localization of multi-location parameters under single / multi-sensor conditions are studied respectively. On this basis, a new precise indoor location method based on RSSI/AOA is proposed. The simulation and test results show that the proposed algorithm can effectively filter the error caused by parameter measurement fluctuation. The accuracy of target location in complex indoor environment is improved effectively. (3) A decision-level fusion algorithm based on strong tracking Kalman filter is proposed. The defects of Kalman filter in moving target location and tracking are solved: poor tracking ability for abrupt state and weak filtering ability for large noise. A fusion tracking algorithm based on exponential fading factor (EFF-STF) is developed by improving the suboptimal fading factor in the strong tracking algorithm and feedback the strong tracking results according to the previous results. The implementation steps of the algorithm are described in detail, and the validity of the algorithm is verified by the establishment of the system's actual data collection, and the superiority of the algorithm in large noise filtering is verified by comparing the performance of the algorithm with the Kalman filter under the same conditions. The positioning error CDF diagram before and after tracking verifies the good correction effect of tracking performance on the positioning error. (4) in this paper, the indoor positioning system based on low power Bluetooth is realized by using MATLAB and VS2010 joint programming, and the positioning results are visualized. The performance of the algorithm is verified. The experimental results show that after information flexible fusion and tracking processing, 85% of the positioning error can be limited to less than 1m, and the requirement of real-time and accurate indoor positioning can be realized.
【学位授予单位】：河南师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN713;TN925
【参考文献】
相关期刊论文 前8条
1 李协;张效义;于旭;胡峗鹏;;相对定位估计的修正克拉美罗界[J];数据采集与处理;2013年02期
2 范馨月;陈庭盈;周非;;基于参数重构的非直达波混合定位算法[J];信号处理;2011年11期
3 孙章国;钱峰;;一种基于指数渐消因子的自适应卡尔曼滤波算法[J];电子测量技术;2010年01期
4 王鑫;赵春晖;戎建刚;;多路延迟结构的修正MUSIC算法频率估计[J];系统工程与电子技术;2009年04期
5 刘海燕;赵宗贵;刘熹;;D-S证据理论中冲突证据的合成方法[J];电子科技大学学报;2008年05期
6 焦磊;邢建平;张军;张璇;赵朝丽;;一种非视距环境下具有鲁棒特性TOA无线传感网络定位算法[J];传感技术学报;2007年07期
7 凌林本,李滋刚,陈超英,谷云彪,李传学;多传感器数据融合时权的最优分配原则[J];中国惯性技术学报;2000年02期
8 翟翌立,戴逸松;多传感器数据自适应加权融合估计算法的研究[J];计量学报;1998年01期
相关博士学位论文 前1条
1 王建辉;基于信息融合的蜂窝网定位技术研究[D];解放军信息工程大学;2011年
相关硕士学位论文 前1条
1 竹博;蜂窝网目标的定位跟踪技术研究[D];解放军信息工程大学;2013年
，本文编号：2438052
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2438052.html
