Neural network indoor ranging method considering signal statistical characteristics
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
万方数据
针对现有神经网络测距方法易陷入局部极值而降低测距精度的问题,提出一种考虑信号统计特征的神经网络蓝牙室内测距方法:提出一种反馈-卡尔曼混合滤波算法进行数据预处理;并分析接收信号的传播特性,引入信号统计特征参数作为输入信号,构建改进麻雀搜索算法优化的埃尔曼(Elman)神经网络(ISSA-Elman)测距模型.实验结果表明,该测距方法能够有效提高测距精度,平均测距误差约为15 cm.
Aiming at the problem that the current neural network ranging method is easy to fall into local extreme value and reduce the ranging accuracy,the paper proposed a bluetooth indoor ranging method by neural network considering signal statistical characteristics:a feedback-Kalman hybrid filtering algorithm for data preprocessing was given;moreover,the propagation characteristics of the received signals were analyzed,the signal statistical feature parameters were introduced as the input signal,and Elman neural network ranging model optimized by improved sparrow search algorithm (ISSA-Elman) was constructed. Experimental results showed that the proposed method could effectively improve the ranging accuracy,with the average ranging error about 15 cm.
received signal strength indication (RSSI)hybrid filteringElman neural networksparrow search algorithmindoor ranging method
NETL
NSTL
万方数据
