基于多维共线度评估机制的WSN网络数据评估算法  

WSN network data evaluation algorithm based on multi-dimensional collinearity evaluation mechanism

作　　者：赵卫[1] Zhao Wei(Information Center, Xianyang Normal University, Xianyang 712000, China)

机构地区：[1]咸阳师范学院信息中心

出　　处：《国外电子测量技术》2018年第3期38-41,共4页Foreign Electronic Measurement Technology

摘　　要：针对当前WSN网络数据评估过程中存在的数据评估收敛性能低,重要数据区分度不强且评估过程极易导致数据拥塞等难题,提出了一种基于多维共线度评估机制的WSN网络数据评估算法。在WSN数据存续周期内进行基于评估预判成本建模,依据WSN网络节点存在的固有经济数据特征进行成本评估;将拥塞率、丢包率等评估参数纳入到评估成本建模过程中,实现了对评估参数的精确建模,解决了当前WSN网络数据评估过程中存在的建模难题;采用高斯评估模式针对耗费（money）、收敛时间（time）、数据稳定率（reliability）进行评估调度函数,且使用柯西准则对评估函数进行优化,从而能够实现对同一时刻的WSN网络的多维数据评估,从而提高了网络的评估效率,降低了数据拥塞现象的发生。仿真实验表明,与超混沌数据评估HCDE算法相比,该算法拥有更低的WSN网络数据拥塞率与更高的数据处理能力,显著改善网络的稳定性能。In order to solve the problem that the data in the current WSN network data evaluation process is low,the important data segmentation degree is not strong and the evaluation process can easily lead to serious data congestion.A WSN based on multi-dimensional collinearity evaluation mechanism is proposed.Network data evaluation algorithm.Firstly,the cost estimation based on the multi-dimensional parameter evaluation is carried out in the WSN data persistence period,and the cost evaluation is based on the inherent economic data characteristics existing by the WSN network node.Secondly,the evaluation parameters such as congestion rate and packet loss rate are included in the evaluation cost.And then the multidimensional function is constructed by using the Gaussian evaluation model for money,convergence time（time）,data stability rate（reliability）,which can realize the multi-dimensional integration data evaluation of WSN network at the same time,and thus greatly Improve the evaluation efficiency of the network,and reduce the occurrence of data congestion.The simulation results show that the proposed algorithm can greatly reduce the data congestion of WSN network and improve the stability of the network,and has good complex environment adaptability and high practical value.

关 键 词：无线传感网络 数据评估 评估预判成本 经济数据特征 高斯评估 柯西准则 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] TN92[自动化与计算机技术—计算机科学与技术]