摘要
为提高基于RSSI定位算法的精度和实时性,提出一种基于nRF51822与LSM303DLHC传感器融合的KNN定位算法。将LSM303DLHC传感器融合至nRF51822芯片中,勘测定位终端相对于参考坐标系的航向角,离线阶段将参考坐标与蓝牙AP接收到的信号聚类值绑定形成指纹库,定位阶段将勘测到的nRF51822芯片RSSI衰减信息与MAC地址信息映射为实时定位向量,采用欧式距离加权KNN算法匹配指纹库得到最终的定位点。实验结果表明,该算法有效补偿了RSSI信号跳变产生的误差,极大提升了定位算法的实时性与精度。
To improve the accuracy and real-time ability of RSSI localization algorithm,a KNN localization algorithm based on nRF51822 and LSM303 DLHC sensor fusion was presented.LSM303 DLHC sensor was integrated into the nRF51822 chip to survey the heading angle of the positioning terminal relative to the reference coordinate system.The fingerprint database was established based on the reference coordinate and the signal clustering value received by Bluetooth AP at offline phase.At positioning phase,the nRF51822 chip RSSI attenuation information and MAC address information were mapped to real-time location vector.The E-distance weighted KNN algorithm was used for matching the fingerprint database to get the final locating point.Experimental results show that the proposed algorithm can effectively compensate the error of RSSI signal hopping,which greatly improves the real-time ability and accuracy of the localization algorithm.
作者
王超
姚瑞玲
ANG Chao 1,YAO Rui-ling 2(1.Publishment and Media Department,Chongqing Business Vocational College,Chongqing 401331,China;2.Light Industrial Engineering Department,Sichuan Technology and Business College,Dujiangyan 611830,Chin)
出处
《计算机工程与设计》
北大核心
2018年第7期1946-1953,共8页
Computer Engineering and Design
基金
重庆市众创空间建设专项基金项目(cstc2015pt-zckj0467)
重庆市高等教育教学改革研究基金项目(153301)
