无线传感网中模糊逻辑分簇和数据融合技术研究

发布时间：2018-08-09 10:55

【摘要】：无线传感器网络(Wireless Sensor Network,WSN)由大量分布在特定区域,具有感知、存储和通信能力的无线节点组成。WSN具有广泛地应用前景。譬如应用于森林防火、智慧小区、智能穿戴、火车站安全监测等。但是受WSN自身特点的限制,节点需要将收集到的数据进行融合处理,降低能耗,提高数据准确性。因此,无线传感器网络数据融合技术成为研究重点之一。本文将节省能量消耗作为基本要求,以提高数据融合准确性和实时性为主要目标,在分簇技术和数据融合技术上做了创新性研究。本文具体内容如下:首先,本文对无线传感器网络的背景现状、体系结构、网络特点,以及无线传感器网络的关键技术——数据融合技术进行概述,并对数据融合算法的分类、性能和局限进行了深入分析。其次,本文通过分析现有分簇技术,提出了一种新的能量均衡的模糊非均匀分簇算法(Energy Enhanced Unequal Fuzzy Clustering Algorithm,EEUFC)。该算法通过计算节点相对密度和到基站的距离随机选取临时簇头;然后引入模糊理论估计竞争半径,并将相对密度与竞争半径、剩余能量一起作为选举最终簇头的参考变量。成簇时,节点根据距离和代价选择簇头,更好地避免了“热区”现象,均衡了能量消耗。再次,本文将模糊逻辑、矩阵加权等思想应用于数据融合技术,考虑了信息收集和传输中对数据准确性和实时性的要求,提出了一种模糊加权的数据融合算法(Fuzzy Weighted Algorithm for Data Fusion,FWADF)。该算法基于分簇模型,考虑了外界因素的影响,在计算可信度的基础上,对收到的数据分别在簇头和基站中进行融合处理,确保为用户提供准确、实时的数据信息。最后,通过使用仿真软件NS-2(Network Simulator-Version 2,NS-2)对本文算法进行了实验仿真。实验结果表明:本文提出的能量均衡的模糊非均匀分簇算法平衡了节点能量且避免了“热区”;提出的模糊加权的数据融合算法提高了数据准确性和实时性,二者共同延长了网络的生命周期。
[Abstract]:Wireless sensor network (WSN) is composed of a large number of wireless nodes with sensing, storage and communication capabilities, which are distributed in a specific area. WSNs have a broad application prospect. For example, used in forest fire prevention, intelligent community, intelligent wear, railway station safety monitoring and so on. However, due to the limitation of WSN itself, nodes need to fuse the collected data to reduce energy consumption and improve the accuracy of data. Therefore, wireless sensor network data fusion technology has become one of the focus of research. In this paper, energy saving is taken as the basic requirement, and the main goal of this paper is to improve the accuracy and real-time of data fusion. The innovative research on clustering technology and data fusion technology has been done. The specific contents of this paper are as follows: firstly, this paper summarizes the background, architecture, network characteristics and the key technology of wireless sensor network data fusion, and classifies the data fusion algorithm. The performance and limitation are analyzed in depth. Secondly, by analyzing the existing clustering techniques, a new fuzzy nonuniform clustering algorithm, (Energy Enhanced Unequal Fuzzy Clustering algorithm, is proposed. By calculating the relative density of the node and the distance from the base station, the algorithm randomly selects the temporary cluster head, and then introduces the fuzzy theory to estimate the competition radius, and takes the relative density and the competition radius together as the reference variables for the election of the final cluster head. In clustering, the cluster heads are selected according to the distance and cost, thus avoiding the "hot zone" better and balancing the energy consumption. Thirdly, this paper applies fuzzy logic and matrix weighting to data fusion technology. Considering the requirement of data accuracy and real-time in information collection and transmission, a fuzzy weighted data fusion algorithm (Fuzzy Weighted Algorithm for Data fusion FWADF is proposed. Based on clustering model and considering the influence of external factors, the received data are fused in cluster head and base station on the basis of calculating credibility, so as to provide accurate and real-time data information for users. Finally, the simulation software NS-2 (Network Simulator-Version 2 / NS-2) is used to simulate the algorithm. The experimental results show that the proposed fuzzy non-uniform clustering algorithm balances node energy and avoids "hot zone", and the proposed fuzzy weighted data fusion algorithm improves the accuracy and real-time performance of the data. The two extend the life cycle of the network together.
【学位授予单位】：辽宁大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【参考文献】