基于KM-FNN的无线传感器网络缺失数据重建算法  

KM-FNN Algorithm for Missing Data Reconstruction in Wireless sensor Networks

作　　者：武加文 李光辉[1,2] WU Jiawen;LI Guanghui(School of IoT Engineering,Jiangnan University,Wuxi Jiangsu 214122,China;Research Center of IoT Technology Application Engineering(MOE),Wuxi Jiangsu 214122 China)

机构地区：[1]江南大学物联网工程学院,江苏无锡214122 [2]物联网技术应用教育部工程技术研究中心,江苏无锡214122

出　　处：《传感技术学报》2019年第8期1245-1252,共8页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目(61174023,61472368);无锡市国际科技研发合作项目(CZE02H1706)

摘　　要：在环境监测领域,无线传感器网络节点受恶劣环境或软硬件故障的影响,在数据传输的过程中不可避免地发生数据丢失的现象,这将严重影响监测数据流的可靠性和用户的科学决策。提出了一种基于自适应的K-均值算法和模糊神经网络相结合的分簇式无线传感器网络缺失数据重建算法(KM-FNN)。该算法使用模糊神经网络模型重建缺失数据,并引入自适应机制适时更新训练模型。针对不同数据集的仿真结果表明,与以往同类算法相比,KM-FNN具有更好的缺失数据重建准确性。In environmental monitoring wireless sensor network,it’s unavoidable that sensor data lose in data transmission due to the hostile environment and hardware or software fault in sensor nodes,which has a serious impact on the reliability of data and scientific decision of the final users.This paper presents a missing data reconstruction algorithm(KM-FNN)in clustered wireless sensor networks based on adaptive K-means algorithm and fuzzy neural network.KM-FNN reconstructs the missing data using the fuzzy neural network model,and introduces an adaptive mechanism to update the model.Simulation results on different real-world datasets show that,compared with other existing methods,KM-FNN has superior performance for reconstructing missing data,and can effectively improve the reliability of sensor data.

关 键 词：无线传感器网络 数据重建 K-均值 模糊神经网络 环境监测 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]