| 一种传感器网络时空融合模型 |
| |
| 引用本文： | 康健,左宪章,唐力伟,张西红,李浩.一种传感器网络时空融合模型[J].军械工程学院学报,2010,22(3):63-66. |
| |
| 作者姓名： | 康健 左宪章 唐力伟 张西红 李浩 |
| |
| 作者单位： | [1]军械工程学院火炮工程系,河北石家庄050003 [2]军械工程学院电气工程系,河北石家庄050003 [3]军械工程学院计算机工程系,河北石家庄050003 [4]63880部队,河南洛阳471003 |
| |
| 基金项目： | 国家自然科学基金资助项目（60672143） |
| |
| 摘 要： | 针对多传感器间存在时空相关性和先验知识未知的情况,提出了一种基于核偏最小二乘最优加权的传感器网络时空融合模型。首先用核偏最小二乘法对每个传感器在不同时刻的测量值进行融合,然后将各个传感器同时刻的估计值进行空间最优融合,从而得到被测参量的最终融合估计。该模型还能排除那些取值远离样本点集合平均水平的奇异点。理论分析和融合实验表明,该模型能取得较高的融合精度,并能显著节省网络能量。
|
| 关 键 词： | 核函数 传感器网络 数据融合 偏最小二乘 方差估计 |
Spatial-temporal Data Fusion for Sensor Networks |
| |
| KANG Jian,ZUO Xian-zhang,TANG Li-wei,ZHANG Xi-hong,LI Hao.Spatial-temporal Data Fusion for Sensor Networks[J].Journal of Ordnance Engineering College,2010,22(3):63-66. |
| |
| Authors: | KANG Jian ZUO Xian-zhang TANG Li-wei ZHANG Xi-hong LI Hao |
| |
| Institution: | 1. Department of Artillery Engineering 2. Department of Electrical Engineering 3. Department of Computer Engineering Ordnance Engineering College, Shijiazhuang 050003, China; 4. PLA Unit 63880,Luoyang 471003, China) |
| |
| Abstract: | To solve the problem that spatial-temporal correlation between the multi-sensors and the prior knowledge is unknown, a new spatial-temporal data fusion model for sensor networks is presented. In this model,the data that is monitored by a sensor at different times are fused first in time order based on kernel partial least squares,and then the temporal fusion data are fused in space based on optimal weighted fusion. The model also can eliminate the overlapping and meaningless information. Theoretical analysis and experimental results show that the model can achieve high precision fusion and flexibility. It also can observably reduce the number of transmissions in sensor networks to save energy. |
| |
| Keywords: | kernel sensor networks data fusion partial least squares variance estimation |
| 本文献已被 维普 等数据库收录！ |
|
