摘要
提出了一种基于非线性支持向量机回归SVR、粒子群优化PSO和卡尔曼滤波KF的室内定位算法。对室内运动的物体,采用SVR训练RSSI和距离,构建RSSI和距离的非线性关系,提高RSSI测距精度;再根据待定位标签到各个阅读器的距离,利用PSO优化算法来确定待定位物体各时间点的位置坐标;最后通过卡尔曼滤波对运动轨迹进行滤波。实验结果表明,该算法提升了室内运动物体的定位精度,减小了定位误差。
In this paper,an indoor localization algorithm is proposed based on nonlinear support vector regression(SVR),particle swarm optimization(PSO)and Kalman filtering(KF).For indoor moving objects,SVR is used to train received signal strength indication(RSSI)and distance,and then the nonlinear mapping relation between the RSSI and the distance is built to enhance the RSSI ranging accuracy.In addition,the PSO optimization algorithm is used to estimate the position coordinates of target objects at each time points according to the distance between target tag and the dif ferent readers.Finally,the KF is used to filter the trajectory.The experimental results show that the proposed algorithm can enhance the localization accuracy of indoor moving objects and reduce the localization error.
作者
杨丽
刘沁舒
徐洁
胡静
宋铁成
蒋宗清
YANG Li;LIU Qinshu;XU Jie;HU Jing;SONG Tiecheng;JIANG Zongqing(School of Information Science and Engineering,Southeast University,Nanjing 211189,China;Wuxi Pin-guan IoT Technology Co.,Ltd.,Wuxi 214000,China)
出处
《移动通信》
2019年第2期78-80,共3页
Mobile Communications
基金
国家自然科学基金(61372104
61771126)
