海底光传感网络对水污染采集数据监测研究  

Monitoring of Data Acquisition of Water Pollution Based on Submarine Optical Sensor Network

作　　者：左名久[1] 

机构地区：[1]海军工程大学电子工程学院,湖北武汉430033

出　　处：《计算机仿真》2018年第2期455-458,共4页Computer Simulation

摘　　要：对海底光传感网络水污染采集数据监测进行监测,能够高效改善海底污染问题,对海底污染进行有效治理。对海底网络水污染采集数据的监测,需要采用模糊隶属度确定每个数据样本所属类别,计算出每组数据的聚类中心,完成数据的高效监测。传统方法计算出数据聚类的初始中心,提取水污染数据辨识阈值,但忽略了计算数据的聚类中心。提出基于二次聚类的海底光传感网络水污染采集数据监测方法。基于相对熵理论计算出网络数据的维度,给出时间窗口对应网络数据源IP地址维度的熵,得到网络数据的层次结构,采用有效指数准则对数据进行聚类,采用模糊隶属度确定每个数据样本所属类别,计算出每组数据的聚类中心,获取数据集的最优模糊分类阈值,由此进行海底光传感网络水污染采集数据的监测。实验证明,所提方法采集数据监测精度高,为提升海底环境监测质量奠定了基础。The aim of this article is to overcome defect of traditional monitoring method for acquired data of water pollution in seabed optical sensing network. Based on secondary clustering, a new monitoring method is proposed. Firstly, based on relative entropy, dimension of network data is worked out and entropy of time window corresponding to IP address dimension of network data source is provided. Then, hierarchical structure of network data is obtained and effective index criterion is used to cluster data. Category of each sample is confirmed using fuzzy degree of mem- bership and cluster center of each group of data is worked out to acquire optimal threshold of fuzzy classification of da- ta set. Thus, the monitoring is completed. Experimental results prove that the method has high monitoring precision. It lays foundation for improving monitoring quality of seabed environment.

关 键 词：光传感网络 水污染 采集数据监测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]