无线电工程2025，Vol.55Issue(3)：558-565,8.DOI:10.3969/j.issn.1003-3106.2025.03.012
一种基于自适应平滑及融合区域特征的室内实时定位方法
An Indoor Real-time Location Method Based on Adaptive Smoothing and Fusion of Regional Features
摘要
Abstract
With the expansion of the subway line network and the complexity of the spatial structure in the station,the daily operation and maintenance of the subway become more complex.The central control room needs to schedule based on the real-time location information of the personnel,so the demand for location services is increasingly strong.Unlike the outdoor space,the environment and structure inside the subway station are more complex and changeable.Affected by inaccessible areas such as elevators and walls,common indoor positioning algorithms have problems such as time delay,point drift,and low accuracy.To solve the above problems,a real-time indoor Long Short Term Memory(LSTM)location method based on adaptive smoothing and fusion of regional features is proposed.Based on the innovative smoothing analysis method,the noise discrimination threshold is adaptively determined for trajectory smoothing.Integrating Bluetooth location information and key area features within the station,the prediction of position coordinates is completed based on LSTM neural network model and the result of the adaptive weighted average of predicted coordinates and measured coordinates is taken as the final position coordinates to provide high-precision location information.Finally,experiments are conducted based on the location data in different scenes in subway stations,and the experimental results show that the location error is reduced by 1.2 m,which verifies the effectiveness of the method.关键词
室内/自适应/关键区域/长短期记忆Key words
indoor/adaptive/critical region/LSTM分类
信息技术与安全科学引用本文复制引用
张艳,赵敏,臧艳军,靳亚宾..一种基于自适应平滑及融合区域特征的室内实时定位方法[J].无线电工程,2025,55(3):558-565,8.基金项目
国家重点研发计划(308项目)National Key R&D Program of China(308) （308项目）
