基于混合高斯模型测距误差修正和EM-SOM的节点定位算法设计

发布时间：2018-03-13 14:01

本文选题：无线传感器网络　切入点：混合高斯模型　出处：《计算机测量与控制》2014年11期 　论文类型：期刊论文

【摘要】：针对RSSI信号强度定位方法中当发射节点和接收节点之间的无线电传播路径被障碍物阻挡而造成噪声误差,使得节点定位欠精确的问题,设计了一种基于混合高斯模型进行测距误差修正和EM-SOM的节点定位算法;首先,通过混合高斯模型对RSSI获得的测量距离误差进行建模,通过EM方法对混合高斯模型中的各参数即各高斯模型的权值、均值和协方差进行训练,采用自组织的SOM对测量距离样本进行聚类,获得各高斯模型的初始权值,最后,将测量距离输入训练后的高斯模型获得较为真实的距离值,并通过极大似然估计进行节点的定位;实验结果表明文中方法能对具有噪声误差的节点进行定位,且相对其它方法,具有平均定位误差和均方根误差小的优点,具有较大的优越性。
[Abstract]:In the RSSI signal intensity localization method, when the radio transmission path between the transmitting node and the receiving node is blocked by obstacles, the noise error is caused, which makes the node location inaccurate. In this paper, a hybrid Gao Si model for ranging error correction and EM-SOM node location algorithm is designed. Firstly, the measurement distance error obtained by RSSI is modeled by the mixed Gao Si model. The EM method is used to train the weights, mean values and covariances of various parameters in the mixed Gao Si model, and the self-organized SOM is used to cluster the measured distance samples to obtain the initial weight values of each Gao Si model. The measured distance is input to the trained Gao Si model to obtain the real distance value, and the node location is carried out by maximum likelihood estimation. The experimental results show that the proposed method can locate the nodes with noise errors, and compared with other methods. It has the advantage of small mean positioning error and root mean square error.
【作者单位】：南通理工学院软件工程系;
【基金】：南通市科技计划项目(BK2012035) 院特色专业建设项目(201104)
【分类号】：TP212.9;TN929.5

【参考文献】