基于蓝牙信标和指纹库匹配的室内定位算法研究

发布时间：2019-06-17 16:44

【摘要】：位置信息、定位技术与位置服务开启了新时代的研究热潮,覆盖到了智能交通,智能家居,智慧工业、农业、商业,智慧城市等诸多领域。GPS和蜂窝网定位技术广泛用于室外位置服务,但由于非视距和多径影响,信号强度和定位精度都无法达到室内定位要求,且耗电速度快,系统成本高。目前的室内定位技术主要是无线定位技术,从近几年的室内定位研究热潮中可以看出,蓝牙低功耗4.0技术在高精度、低功耗、易部署、系统简单、成本低上都较有优势。同时智能手机、iPhone、iPad等智能终端设备的快速发展,且大多都支持BLE功能,更加促进了室内i Beacon技术的应用。可以说,蓝牙定位技术将会成为室内定位技术的一大支柱,前景广阔。本文在室内定位的定位技术与定位算法的研究现状基础上,深入分析研究了基于蓝牙信标iBeacon的指纹库匹配定位,分析了选用iBeacon指纹库相关性匹配定位的可行性及优势。本文的主要研究工作如下:(1)对典型室内办公室环境的iBeacon信标布置进行研究,为了使定位区域中各个位置处采集的RSSI序列有明显区分,且结合实际布置成本,及定位精度要求,确定了3~5米之间布置一个iBeacon信标基站的信标布置方案。(2)对实验环境内RSSI采集的方向、时间、人员干扰进行了重点研究分析,并对参考点的分布做了规划,确定了单点多方向多次采集方案。采集指纹库之后,对其进行融合卡尔曼均值滤波处理,构建稳健较准确的iBeacon信号指纹库。(3)提出了相关性匹配指纹库算法,对指纹库进行两次筛选后,求解未知点与参考点的相关系数γ,选出在0.8?|??(27)(16)范围内的进行显著性检验,得到K个匹配性较高的指纹库参考点,以相关系数绝对值为加权系数对参考点坐标进行加权平均得到预估位置。实验结果表明,相关系数匹配位置指纹库算法可将定位误差在2米以内的概率从65%提高到92%,对比于常用的KNN匹配定位算法具有定位精度高、定位时间短、算法稳定等优势。(4)对基于iBeacon信标的室内定位系统的指纹库采集和实时定位两部分进行了设计与实现,确定了Android移动端平台,负责采集处理iBeacon信号,并上传iBeacon信号指纹;同时在J2EE架构的后台服务器端,保存位置指纹库进行相关性匹配实时定位,然后反馈给用户端,实现相关性匹配定位结果的显示测试。
[Abstract]:Location information, location technology and location service opened a new era of research upsurge, covering many fields such as smart traffic, smart home, smart industry, agriculture, commerce, and smart city. GPS and cellular network positioning technology is widely used for outdoor location services, but due to the non-sight distance and multi-path influence, the signal strength and the positioning accuracy can not reach the indoor positioning requirement, and the power consumption speed is high, and the system cost is high. The present indoor positioning technology is mainly the wireless location technology, which can be seen from the indoor positioning research upsurge in recent years, and the Bluetooth low power consumption 4.0 technology has the advantages of high accuracy, low power consumption, easy deployment, simple system and low cost. At the same time, smart terminal devices such as smart phone, iPhone, iPad and other intelligent terminal devices have developed rapidly, and most of them support the BLE function, and the application of the indoor i Beacon technology is more promoted. It can be said that the Bluetooth positioning technology will become a big pillar of the indoor positioning technology, and the prospect is wide. In this paper, on the basis of the research status of the positioning technology and the location algorithm of the indoor location, the matching and location of the fingerprint library based on the Bluetooth beacon iBeacon is analyzed, and the feasibility and the advantage of the correlation matching and positioning of the iBeacon fingerprint library are analyzed. The main research work of this paper is as follows: (1) The iBeacon beacon arrangement of the typical indoor office environment is studied, and in order to make the RSSI sequence collected at each position in the positioning area to be clearly distinguished, and the actual layout cost and the positioning accuracy requirement are met, And a beacon arrangement scheme of an iBeacon beacon base station is arranged between 3 and 5 meters. (2) The direction, time and personnel interference of the RSSI acquisition in the experimental environment are analyzed and analyzed, and the distribution of the reference point is planned, and the multi-direction multi-direction acquisition scheme is determined. After the fingerprint library is collected, a robust and accurate iBeacon signal fingerprint library is constructed. (3) The correlation coefficient of the unknown point and the reference point is calculated and the correlation coefficient between the unknown point and the reference point is calculated. |? (27) carrying out significance test in the range of (16) to obtain a fingerprint library reference point with high K matching property, and weighting the reference point coordinate with the absolute value of the correlation coefficient as a weighting coefficient to obtain an estimated position. The experimental results show that the correlation coefficient matching position fingerprint library algorithm can increase the probability of the positioning error within 2 meters from 65% to 92%, and compared with the conventional KNN matching and positioning algorithm, it has the advantages of high positioning accuracy, short positioning time and stable algorithm. and (4) designing and implementing the fingerprint database acquisition and real-time positioning of the indoor positioning system based on the iBeacon beacon, And the stored position fingerprint library carries out correlation matching real-time positioning, and then feeds back to the user end to realize the display test of the correlation matching and positioning result.
【学位授予单位】：重庆理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925

【参考文献】