WiFi指纹定位及跟踪技术研究

发布时间：2018-04-05 19:15

本文选题：WiFi指纹定位　切入点：边缘重叠法　出处：《大连理工大学》2014年硕士论文

【摘要】：近些年,随着WiFi技术在无线网络应用中的普及,室内WiFi定位成了研究的热门课题。一种常用的室内定位方法为基于接收信号强度(Received Signal Strength, RSS)的位置指纹法,该方法可行性强、定位成本低,受到研究人员的广泛关注。位置指纹法虽然能够较好地在室内实现精确定位,但是当指纹数量增多时,定位算法的复杂度也会增大,系统的实时性就比较差,尤其是对目标进行跟踪时,对定位系统的实时性要求会更高,传统定位方法难以满足要求。为此,本文较为深入地探讨了基于接收信号强度的定位和跟踪问题,主要工作如下：首先,在实际环境搭建了实验平台,实际测量了大量的数据,分析了影响室内WiFi定位的关键因素,包括RSS与位置的关系、人对RSS的影响、可检测到的AP数目的变化以及RSS的概率分布情况等,为进一步研究提供了保障；其次,为了降低算法的复杂度,本文采用分层定位的思想,将定位分为粗定位和精定位两步。这一方法中,区域划分很重要,本文提出了一种实际可行的边缘重叠法,并分别采用压缩感知和最小欧几里德距离方法进行区域判定。本文在仿真场景和实际环境中进行了实验,将这种方法与现有的经典的聚类方法(比如,仿射传播聚类)做了比较研究,实验结果表明,采用边缘重叠法能达到更好的定位精度；最后,对跟踪技术进行了研究,主要研究了粒子滤波算法,并将基于边缘重叠法的分层指纹定位技术与现有的loose-coupling PF相结合,实现了对目标的跟踪。分层技术的引入,并没有对定位精度造成很大影响,而且减少了算法的复杂度,提高了系统的实时性。
[Abstract]:In recent years, with the popularity of WiFi technology in wireless network applications, indoor WiFi positioning has become a hot topic.A common indoor location method is based on received Signal received Signal fingerprint (RSS). This method is feasible and low in cost, and has been widely concerned by researchers.Although the location fingerprint method can achieve accurate location in the room, when the number of fingerprints increases, the complexity of the location algorithm will also increase, and the real-time performance of the system is relatively poor, especially when tracking the target.The real-time requirement of the positioning system will be higher, the traditional positioning method is difficult to meet the requirements.Therefore, this paper deeply discusses the localization and tracking based on the received signal strength. The main work is as follows:First of all, the experimental platform is built in the actual environment, and a large number of data are measured, and the key factors affecting indoor WiFi positioning are analyzed, including the relationship between RSS and location, the influence of people on RSS.The change of the number of AP and the probability distribution of RSS provide the guarantee for further research.Secondly, in order to reduce the complexity of the algorithm, this paper adopts the idea of hierarchical localization, and divides the location into coarse location and fine location.In this method, the region division is very important. In this paper, a practical edge overlap method is proposed, and the compression sensing method and the minimum Euclidean distance method are used to determine the region.In this paper, experiments are carried out in the simulation scene and the actual environment. The results show that this method is compared with the existing classical clustering methods (for example, affine propagation clustering).The edge overlap method can achieve better positioning accuracy.Finally, the tracking technology is studied, the particle filter algorithm is mainly studied, and the hierarchical fingerprint location technology based on the edge overlap method is combined with the existing loose-coupling PF to achieve the target tracking.The introduction of stratification technology does not have a great impact on the positioning accuracy and reduces the complexity of the algorithm and improves the real-time performance of the system.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【参考文献】