无线传感网络信息分段融合点提取方法仿真

摘要

In order to solve the current situation that a lot of node energy is used to transmit redundant data in wireless sensor networks,this article focuses on a method of extracting information segmentation fusion point in wireless sensor network based on BP neural network.The feature of segmented space of fusion object data set in wireless sensor network is extracted.The eigen matching is used to design overall model of information fusion in wireless sensor network.Then,the network information feature segmented fusion is introduced to locate the spectral function of node and describe the efficiency of information fusion.Based on linear differential fusion,the information feature is fused piecewise to obtain the proportionate score in the task assignment link of information segmentation fusion and describe the structural feature of information segmentation fusion.Neural network is used to describe the cluster structure in wireless sensor network.According to TEEN threshold,nonessential fusion structural features are filtered.The performance function of neural network is used to deal with massive structural feature data of information segmentation fusion in the process of extracting information segmentation fusion nodes,and forward to fusion nodes.Simulation results show that the proposed method can reduce the energy consumption in network and improve the lifetime of wireless sensor networks in the process of extracting information segmentation fusion points.%为了解决无线传感网络普遍存在的节点能量大量用于传输冗余数据的现状,现行的信息融合点提取方法通过建立无线传感网络分布结构模型进行数据结构分析,对冗余数据自适应归集合并,此过程增加了不必要的数据传输,导致节点能量消耗过大、融合点提取时间较长、网络寿命较短.提出一种基于BP神经网络的无线传感网络信息分段融合点提取方法.对无线传感网络融合对象数据集提取分段空间特征,采用本征匹配方法进行无线传感网络信息融合总体模型设计,引入网络信息特征分段融合定位节点谱函数描述信息融合效率,根据线性差分融合进行信息特征分段融合,获得信息分段融合任务分配链路中的比例分值,描述信息分段融合的结构特征.采用神经网络描述无线传感网络簇结构,根据TEEN阈值过滤非必要的融合结构特征,在信息分段融合点提取过程中运用神经网络功能函数处理大量信息分段融合结构特征数据,并转发至融合节点.实验结果表明,所提方法在信息分段融合点提取过程中,在降低网络能耗的同时提升了无线传感网络的生存寿命.