基于DGMM混合滤波与神经网络的室内测距算法  

Indoor Ranging Algorithm Based on DGMM Hybrid Filter and Neural Network

作　　者：蒋羽兴 JIANG Yuxing(Jiangsu University of Science and Technology,Zhenjiang 212000)

机构地区：[1]江苏科技大学常州技师学院,镇江212000

出　　处：《计算机与数字工程》2025年第1期124-126,151,共4页Computer & Digital Engineering

摘　　要：典型的基于接收信号强度指示(RSSI)测距定位方法存在精度较低且不稳定的问题,而单一滤波器存在各种缺陷,为此论文提出一种混合滤波与神经网络结合的测距算法。通过混合滤波处理数据,采用神经网络对测距模型进行非线性拟合。实验结果证明,经过混合滤波与神经网络拟合计算后,能有效降低测距误差。Typical received signal strength indication(RSSI)ranging location method has the problem of low accuracy and in⁃stability,and a single filter has various defects,so this paper proposes a hybrid filtering and neural network combined ranging algo⁃rithm.The data is processed by hybrid filtering,and the ranging model is fitted nonlinear by neural network.Experimental results show that the range error can be effectively reduced after the hybrid filtering and neural network fitting calculation.

关 键 词：加权混合滤波 路径损耗 RSSI测距 神经网络 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]