多传感器辅助的WiFi信号指纹室内定位技术

近年来，基于室内定位的应用服务越来越普及，吸引了大量的研究工作.其中，基于WiFi信号指纹的室内定位技术发展尤为迅速.但无线信号传输易受环境影响，会导致WiFi信号指纹定位存在偏差.为了提高定位精度并减小环境因素带来的不利影响，提出了智能手机内置传感器辅助WiFi信号指纹定位的方法，即利用智能设备上内置的传感器如加速计、陀螺仪等采集数据，计算得到用户轨迹信息和轨迹可信度，将轨迹信息与信号指纹信息结合起来建立综合概率模型，进行位置匹配，确定最近参考点.实验结果表明，与经典WiFi信号指纹定位方法相比，利用传感器所估测的用户轨迹信息能够有效应对环境变化的影响，提高平均定位精度.

Multi-sensor Assisted WiFi Signal Fingerprint Based Indoor Positioning Technology

Indoor positioning is fundamental to many smartphone applications, attracting a great deal of research efforts in recent years. Among them, RSSI (received signal strength indication) based fingerprinting has become an increasingly popular technique for realizing indoor smartphone positioning using existing WiFi infrastructures. However, wireless signal transmission is easily affected by the environment, which may result in the deviation of WiFi signal fingerprint positioning. To address this issue, multi-sensors assisted WiFi fingerprinting method is proposed to improve the performance of RSSI fingerprinting. In this method, the data obtained from the smartphone''s built-in sensors like accelerometer and gyroscopeis is utilized to estimate user''s trajectory along with its credibility. Then a probability model which combines the RSSI fingerprint and users'' trajectory is established to implement a match between fingerprint and location. Experiments show that compared with the classical fingerprint-matching algorithm, the proposed method can effectively reduce the adverse effects of environmental changes on positioning and improve average localization accuracy by making use of the sensor data.