基于蓝牙Mesh-SVM算法的室内定位系统

Wireless Positioning Systems Based on Intelligent Devices of Bluetooth Mesh

随着低功耗蓝牙的更新换代(BLFA.2),蓝牙技术在智能家居、智能设备开发中日显突出。本文针对蓝牙信号多径、易衰弱、穿透能力低等缺陷,采用支持向量机(SVM)神经网络算法来实现系统的室内定位,通过RBF径相基函数为核函数(高斯核)可控的将低维线性不可分数据映射到高维模型中,计算超平面并完成数据分类;基于空间的不可瞬变性,采用反馈滤波对验证结果进行筛选,实现精确定位。实验结果中表明算法定位精度达到1m,与传统算法相比有效的提高了效率与精度。

With low power consumption bluetooth upgrades（BLE4.2）,bluetooth technology,which in the field of smart home and intelligent device development is becoming increasingly prominent.This article directing at the bluetooth signal multipath and weak low penetration defects,try to use support vector machine（SVM） algorithm of neural network to realize the indoor system positioning,through RBF diameter basis function as the kernel function（gaussian kernel） controllable to make lower dimensional linear inseparable data mapped to high-dimensional model,calculate hyperplane and finish data classification;Based on the instantaneous degeneration of the space,the feedback filter is adopted to verify the results and realize accurate positioning.The results of the experiment show that the precision of the algorithm achieve to locate 1 m,which effectively improve the effectiveness and efficiency compared with the traditional algorithm.