基于最邻近相关系数的指纹室内定位新算法

发布时间：2018-01-14 06:13

本文关键词：基于最邻近相关系数的指纹室内定位新算法　出处：《西北工业大学学报》2017年04期 　论文类型：期刊论文

【摘要】：室内定位技术可提供准确的位置信息的服务,因而在多种领域中得到广泛应用。在Wi-Fi网络环境下,相关系数法和KNN算法是基于指纹数据库的2种常用定位算法,但两者的定位精度都十分有限,不能满足室内精确定位的要求。为此提出一种最邻近相关系数算法,该算法结合了均值的相关系数与KNN算法,在一定的室内环境下共同发挥了两者的定位优势,并且较好解决了斯皮尔曼等级相关系数和皮尔逊相关系数的异值点问题。在实际物理空间建立起指纹数据库,选择多个测试点对新算法进行了性能测试,测试结果表明,文中提出的最邻近相关系数法在定位精度上有所提升,分别比相关系数法和KNN法的定位精度提升了38.86%和23.35%,同时在数据运算量上并没有增加,可以在室内定位中广泛运用。
[Abstract]:Indoor positioning technology can provide accurate location information services, so it has been widely used in many fields. In the Wi-Fi network environment. Correlation coefficient method and KNN algorithm are two common localization algorithms based on fingerprint database, but the accuracy of both is very limited. It can not meet the requirements of accurate indoor location. Therefore, this paper proposes a nearest neighbor correlation coefficient algorithm, which combines the mean correlation coefficient with the KNN algorithm. In a certain indoor environment, the location advantages of both are brought into play together, and the problem of the abnormal value of Spelman grade correlation coefficient and Pearson correlation coefficient is solved well, and the fingerprint database is established in the actual physical space. Several test points are selected to test the performance of the new algorithm. The test results show that the nearest neighbor correlation coefficient method proposed in this paper has improved the positioning accuracy. Compared with the correlation coefficient method and the KNN method, the positioning accuracy is improved by 38.86% and 23.355.At the same time, the data operation is not increased, so it can be widely used in indoor positioning.
【作者单位】：西北工业大学电子信息学院;
【基金】：国家自然科学基金(61401360) 中央高校基本科研业务费专项资金(3102017zy026)资助
【分类号】：TN92
【正文快照】： 随着对位置服务的需要,基于个人或是物体的定位技术不断涌现,GNSS(global navigation satellitesystem)在室外已经可以较为精确地解决这个问题,例如GPS(global positioning system)、北斗导航系统和伽利略(Galileo)定位系统等。而在人们活动时间更为长久的室内环境中,GNSS信号

【相似文献】