基于Wi-Fi的室内定位算法研究与实现

发布时间：2018-02-11 17:42

本文关键词： Wi-Fi 定位位置指纹传感器　出处：《武汉理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着智慧城市建设的兴起和移动互联网的快速发展，人们对基于位置的服务需求日益增强，这就要求其能够实现准确的定位与跟踪。目前，定位技术广泛应用于各个领域，而基于GPS的室外定位的成功应用也激励了室内定位系统的研究与开发。然而由于接受信号微弱，GPS系统不能有效运用于建筑物内部和密集的都市区域。目前的室内定位系统采用的技术主要有计算机视觉、红外线、射频识别、超宽带、无线传感网络、AGPS等，但这些系统往往需要部署额外的设施，应用时大受限制。基于Wi-Fi的室内定位系统能够充分利用现有的基础设施，在不需要部署额外设备的情况下，应用于提供室内位置服务。而且智能手机也都内置了Wi-Fi模块，，这使得基于Wi-Fi的室内定位成为可能。所以使用智能手机利用遍布建筑物内的Wi-Fi信号定位成为了一种极具潜力的室内定位技术。本文在对Wi-Fi室内定位技术进行充分研究的基础上，针对现有的基于指纹的Wi-Fi定位算法存在的不足之处提出了相应的改进算法，并设计和实现了Wi-Fi室内定位系统。首先，本文对Wi-Fi的接收信号强度特性和影响因素进行分析，包括RSSI的概率分布，RSSI与距离的关系，人体方位的影响和不同设备的影响。其次，从定位系统构建及工作流程出发，依次探讨了数据采集构建位置指纹数据库，实时定位阶段的预处理过程，定位参考AP选择，信号距离，近邻选取和定位结果计算等。在定位结果计算阶段，提出最密集近邻算法，通过比较可知，相比传统质心法能得到更好的定位精度。然后，对卡尔曼滤波器在动态定位追踪中的作用进行分析。通过手机传感器判断用户运动状态发生变化，使用不同的卡尔曼参数，有效改善了动态定位追踪的效果。接着，通过研究基于手机传感器的定位，提出Wi-Fi定位与传感器定位的融合算法，在无法进行Wi-Fi定位时使用传感器定位，可以有效弥补Wi-Fi定位的不足。同时在Wi-Fi正常定位过程中，能够对Wi-Fi定位波动较大的情况进行较好地校正，有效改善动态定位追踪的效果。最后，设计并实现基于Android手机平台的位置指纹Wi-Fi室内定位系统，并对软件各个模块进行分析。实验结果表明，本文提出的算法能够更有效地实现高精度的Wi-Fi室内定位。
[Abstract]:With the rise of intelligent city construction and the rapid development of mobile Internet, people's demand for location-based services is increasing, which requires them to achieve accurate positioning and tracking. At present, positioning technology is widely used in various fields. The successful application of outdoor positioning based on GPS also encourages the research and development of indoor positioning system. However, because of the weak reception signal, GPS system can not be effectively used in the interior of buildings and in dense urban areas. The main technologies used in the positioning system are computer vision, Infrared, radio frequency identification, ultra-wideband, wireless sensor networks, AGPS, etc., but these systems often need to deploy additional facilities and are severely restricted in their application. Indoor positioning systems based on Wi-Fi can take full advantage of existing infrastructure. It is used to provide indoor location services without the need to deploy additional devices, and smartphones are built into Wi-Fi modules. This makes indoor positioning based on Wi-Fi possible, so using smart phones to use Wi-Fi signals all over the building has become a potential indoor location technology. Based on the research of Wi-Fi indoor location technology, this paper puts forward the corresponding improved algorithm for the existing fingerprint based Wi-Fi localization algorithm, and designs and implements the Wi-Fi indoor positioning system. In this paper, the characteristics and influencing factors of the received signal intensity of Wi-Fi are analyzed, including the relation between the probability distribution of RSSI and distance, the influence of human body orientation and different equipments. Data acquisition and construction of location fingerprint database, preprocessing process of real-time positioning stage, selection of location reference AP, selection of signal distance, selection of nearest neighbor and calculation of location result are discussed in turn. This paper proposes the densest nearest neighbor algorithm, which can get better positioning accuracy than the traditional centroid method. The function of Kalman filter in dynamic location tracking is analyzed. The mobile phone sensor is used to judge the change of user's motion state, and different Kalman parameters are used to effectively improve the effect of dynamic location tracking. By studying the location of mobile phone sensor, the fusion algorithm of Wi-Fi location and sensor location is proposed. When Wi-Fi positioning cannot be carried out, using sensor positioning can effectively compensate for the deficiency of Wi-Fi location. At the same time, in the process of Wi-Fi normal positioning, It can correct the large fluctuation of Wi-Fi location, and improve the effect of dynamic location tracking. Finally, the location fingerprint Wi-Fi indoor positioning system based on Android mobile platform is designed and implemented. The experimental results show that the algorithm proposed in this paper can achieve high precision Wi-Fi indoor positioning more effectively.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【参考文献】