基于蚁群算法的支持向量机室内蓝牙标定定位

针对不同型号的智能手机之间存在硬件差异,导致在使用不同的智能手机进行室内定位时,采集同一蓝牙信号强度观测值存在偏差而影响定位精度的问题,提出了一种基于蚁群算法的支持向量机室内定位蓝牙标定方法.由于支持向量机参数的选取对其预测精度影响较大,因此利用蚁群算法避免人为盲目地选择支持向量机的参数,优化模型并提高预测精度.实验结果表明:基于蚁群算法的支持向量机室内定位蓝牙标定方法相比标定前的精度提高了37.3%,可以有效地进行室内定位.

The Indoor Positioning Method with Bluetooth Calibration of Supporting Vector Machine Based on Ant Colony Algorithm

For hardware differences between different models of smartphones,it can occur a deviation when smartphone indoor positioning is used to collect the same bluetooth signal strength measurements,it may affect the positioning accuracy.The indoor positioning bluetooth calibration method of support vector machine(SVM)based on ant colony algorithm is proposed in this paper.Since the selection of SVM parameters has a great influence on its prediction accuracy,ant colony algorithm can avoid blindly selecting SVM parameters,optimize the model and improve the prediction accuracy.Experiments show that the accuracy of the proposed method is 37.3% higher than that before calibration,and the proposed method can effectively perform indoor positioning.