| 摘要: |
| 基于WiFi的室内定位具有低成本、易部署、覆盖范围广、精度高等优点，成为室内定位技术研究的热点.WiFi指纹定位法可以对抗多径的影响，具有较高的定位精度.但是由于智能移动终端的种类繁多，使得室内定位系统离线建立的指纹数据库难以兼容不同的智能移动设备，降低了定位系统的适用性.为解决此问题，提出了加权余弦相似度算法，使用信号强度的加权余弦相似度作为匹配特征.实测表明，采用加权余弦相似度算法可有效解决终端差异性，提高了室内定位系统的精度和普适性. |
| 关键词: 室内定位 终端差异 加权余弦相似度 WiFi |
| DOI：10.3969/J.ISSN.1000-5137.2017.01.021 |
| 分类号: |
| 基金项目:国家自然科学基金项目（61571135） |
|
| WiFi fingerprint indoor localization based on weighted cosine similarity |
|
Wu Yun1, Jiang Xinhua2, Qi Yanfei1, Jiang Xueqin1
|
|
1.College of Information Science & Technology, Donghua University, Shanghai 201620, China;2.Rail Transit Division, Shanghai Aerospace Power Technology, Co., Ltd., Shanghai 201615, China
|
| Abstract: |
| Benefiting from the low cost,easy deployment,wide coverage and high precision,the WiFi based indoor positioning has attracted a lot of research attentions recently.The fingerprint location method has high location accuracy since it can combat the multipath effect.But because of the wide variety of intelligent mobile terminals,it is difficult to be compatible with different intelligent mobile devices,which reduces the universality of the positioning system.In order to overcome this shortcoming,a weighted cosine similarity algorithm was proposed.The experiments demonstrate that the proposed algorithm can effectively solve the terminal difference and improve the accuracy and universality of the indoor positioning system. |
| Key words: indoor positioning terminal difference weighted cosine similarity WiFi |
