无线传感器网络误删数据自匹配追踪仿真研究

Simulation Study on Self-matching Tracking of Error Deleted Data in Wireless Sensor Networks

针对基于矩阵补全技术、基于事件序列检测以及基于最近邻的三种追踪方法对于误删的关键无线传感数据,存在不能及时进行追踪,性能不足的问题,提出一种无线传感器网络误删数据自匹配追踪方法。通过采集感知对象的信息,获取采样矩阵及传感器网络信号,实现对基准数据集的提取;分解K-稀疏的信号,进行数据信号稀疏变换;经过稀疏变换的数据信号,使用追踪算法对误删信号进行高概率的追踪,构建数据信号匹配矩阵;根据得到的结果实现误删数据追踪。结果表明:与基于矩阵补全技术、基于事件序列检测法以及基于最近邻的三种追踪方法相比,利用匹配追踪方法进行无线传感器网络误删数据追踪后,与原始信号匹配度达到98.7%,完整性更高、误差更小,解决了三种传统追踪方法性能不足的问题,能有效减少关键数据丢失,保障监测结果的准确性。

In this article, a self-matching tracking method for the erroneously deleted data in wireless sensor network was put forward. Firstly, we got the sampling matrix and sensor network signal by collecting the information of perceptual object, and thus to extract the benchmark data set. Secondly, we decomposed K-sparse signal and performed the sparse transformation of data signal. Thirdly, we used the tracking algorithm to track the erroneously deleted signal in a high probability way based on the data signal after sparse transformation, and then constructed a data signal matching matrix. Finally, we could track the erroneously deleted data. Following conclusions can be drawn from the results. The proposed method is compared with the method based on the matrix complement technology, the method based on the event sequence detection and the method based on the nearest neighbor. After tracking the erroneously deleted data in wireless sensor network by the matching tracking method, the matching degree reaches 98.7% with the original signal. The integrity is higher and the error is smaller. The problem of insufficient performance existing in the three traditional tracking methods is solved. Meanwhile, the loss of key data can be effectively reduced and the accuracy of the monitoring result is improved.