基于GF-KF修正RSSI的室内指纹定位方法

针对Wi-Fi信号易受噪声等外界不确定因素的影响以及移动终端接收信号强度指示（RSSI）与真实值存在偏差而导致定位精度不高的问题,本文提出了一种基于GF-KF修正RSSI的室内指纹定位方法．由于采集的RSSI不稳定,该方法利用RSSI类高斯分布的特性,对RSSI数据进行高斯拟合,以得到较为确定的RSSI值．在此基础上,引入卡尔曼滤波算法对拟合后的RSSI数据进行误差修正,结合加权K近邻（WKNN）匹配算法进行定位．实验结果表明:本文方法的平均定位误差为1.5 m,2.0 m以内的误差累积分布概率为90.06％,定位效果优于同类方法． 

Indoor fingerprint positioning method based? on RSSI modified by GF-KF

Aiming at the problems that Wi-Fi signals are susceptible to external uncertainties such as noise, and the RSSI received by mobile terminals deviates from the true value, which results in low positioning accuracy, this paper proposes an indoor fingerprint positioning method based on RSSI modified by GF-KF. Because the collected RSSI is unstable, this method uses the characteristics of the RSSI like Gaussian distribution to perform a Gaussian fit on the RSSI data to obtain a relatively determined RSSI value. Based on this, a Kalman filter algorithm is introduced to correct the RSSI data after fitting, and the WKNN matching algorithm is used to locate. The experimental results show that the average positioning error of the method in this paper is 1.50 m, and the cumulative distribution probability of positioning errors within 2.0 m is 90.06％, and the positioning effect is better than similar methods.