基于RSSI利用KNN位置指纹法的室内定位（卡尔曼滤波）

%Organized by GJH_SJTU %代码使用方法： %主函数：Rssi_localization_KF.m %其余函数均为运行Rssi_localization_KF.m时调用的函数 %利用RSS进行室内定位仿真实验的程序步骤：（对应Rssi_localization_KF.m的代码步骤） 1、通过射线跟踪技术，构建20x15x4m空间内的RSS指纹数据库； 2、在RSS指纹数据库中抽取离线数据集； 3、利用抽取出来的数据集作为KNN算法的训练数据集； 4、在空间中，随机产生一条移动机器人的运动轨迹； 5、采集移动机器人的运动轨迹点处的RSS值； 6、利用KNN算法识别得到运动轨迹坐标； 7、利用卡尔曼滤波对运动轨迹进行滤波； 8、比较滤波前后的定位精度变化。 %How to use the code: %The main function: Rssi_localization_KF.m %The remaining functions are all functions that are called when running Rssi_localization_KF.m The program steps of using RSS to perform indoor positioning simulation experiments: (corresponding to the code step of Rssi_localization_KF.m) 1. Construct an RSS fingerprint database in a 20x15x4m space by using ray tracing technology; 2. Extract offline data sets in the RSS fingerprint database; 3. Use the extracted data set as the training data set of the KNN algorithm; 4. In the space, randomly generate a movement trajectory of a mobile robot; 5. Collect the RSS value at the movement track point of the mobile robot; 6, using KNN algorithm to identify the motion trajectory coordinates; 7. Kalman filter is used to filter the motion trajectory; 8. Compare the positioning accuracy changes before and after filtering.