面向水质监测的无线传感器网络能效优化与综合评估研究
发布时间：2018-08-19 06:02
【摘要】：无线传感器网络(Wireless Sensor Network, WSN),即随机分布于监测环境中的传感器节点通过无线通信的方式,协同工作并构成网络,将侦听到的有效信息经过计算和融合等简单处理,再发送至基站。微传感器、低功耗电子技术和射频通信技术的发展推动了WSN的产生和蓬勃发展。虽然WSN以低功耗和灵活自组织的特点,取得广泛的应用前景,然而WSN的应用环境大多是复杂多变、不宜部署有线网络的工程应用和环境监测,要求节点采用微型化设计,并使用电池供电。电池的能量极为有限,且部署之后再次补充电能极为不易,故在保证网络覆盖质量和通信要求的前提下减少能耗,提高网络的能量效率,延长节点和网络的生存时间,是无线传感器网络研究的重要内容。 WSN的能量消耗可以分为“必要能耗”和“非必要能耗”。其中必要能耗有三个用途：(ⅰ)发送与接收数据；(ⅱ)处理查询请求；(ⅲ)向邻居节点转发查询或数据包。“非必要能耗”是用以：(ⅰ)空闲侦听；(ⅱ)数据包冲突导致的重发；(ⅲ)侦听；(ⅳ)产生或处理控制包。网络节省能量的方式有减少“非必要能耗”的,也有以减少“必要能耗”为手段的。这些节能方法可以分为三类：第一类,高能效路由技术；第二类,数据处理技术；第三类,拓扑控制技术。这些方法都是节约能量、提高网络能效性能的有效方法,但是仍有许多理论障碍技术瓶颈尚未解决。 本文在“基于传感器网络的水质在线检测装置”等应用需求下,针对无线传感器网络在节能技术方面的挑战,在综合分析WSN分簇路由原理与方法的基础上,建立能耗均衡的分簇路由协议,以延长网络生存时间；针对水质监测与评价的应用环境,研究了无线传感器网络数据融合的策略与方法,提出了一种高能效的基于分簇路由的数据融合策略和基于变异系数的模糊综合指数数据融合算法；最后总结若干描述能效的概念,提出了无线传感器网络分簇协议的能量效率综合评估模型与方法。论文的主要内容及创新性成果包括： (1)分簇是无线传感器网络节约能耗、提高能效的一种有效途径,传统的分簇路由协议的簇头选择,大多是产生随机数并与阈值比较的方式实现,这种方式导致无线传感器网络簇头选择的随机性、簇头数目和性能参数的浮动性。本文以LEACH等分簇协议为基础,摒弃随机数与阈值比较的簇头选择方法,提出两种基于报告包的分簇协议：PNSCH和PCHSIF协议。PNSCH协议将节点剩余能量信息在每轮结束时以报告包形式发送至当前簇头,并选择簇内剩余能量最大的节点作为下一轮的簇头。PCHSIF协议将节点剩余能量和距基站距离作为簇头判定的依据,并将其发送至当前轮的簇头,以备下一轮簇头判定使用,且簇头数目调整为最优值。这两种协议在簇的建立过程中,成员节点和簇头节点之间会进行报告包发送和接收,虽然消耗一定的能量,但是可以达到平衡网络能耗、节约网络整体能耗的目的。 (2)减少数据通信量也是提高能耗的有效途径,数据通信量与传感器采集信号的信息量大小密切相关,这些信息存在一定的冗余。数据融合技术可解决数据冗余问题。为此,研究工作综合监测区域整体水质评价的问题,研究多传感器的数据融合策略和算法：提出了三种应用于随机部署环境的无线传感器网络数据融合策略,并分析了三种策略在能效方面的性能；提出了一种面向整个监测区域水质综合评价的数据融合算法,该算法以模糊理论为基础,通过变异系数法确定不同水质指标的权重,并设计了一种加权综合平均算法以弱化局部特异数据对整个监测区域的综合评价的影响,最终以模糊综合指数表征局部区域的水质综合评价类别,以及整个区域的水质综合评价类别。 (3)首次提出了无线传感器网路分簇路由协议能效综合评估的概念,研究了能效参数体系,并将多目标决策理论引入无线传感器网络分簇路由协议的能效评估的研究之中。将同趋势和归一化处理与TOPSIS算法结合,作为综合评估的计算方法。并以LEAC、LEACH-C、SEP和HEED分簇协议为例,验证了能效综合评估方法在三方面的效果：相同部署条件下的高能效分簇路由协议选择、同一分簇路由协议的高能效部署方案选择,以及不同部署条件下的不同分簇路由协议的高能效方案选择。
[Abstract]:Wireless Sensor Network (WSN), which is distributed randomly in the monitoring environment, works together and forms a network by means of wireless communication, and sends the effective information to the base station after simple processing such as calculation and fusion, micro-sensor, low-power electronic technology and radio frequency communication technology. The development of WSN promotes the emergence and vigorous development of WSN. Although WSN has a wide application prospect because of its low power consumption and flexible self-organization, the application environment of WSN is mostly complex and changeable, so it is not suitable to deploy the engineering application and environmental monitoring of wired network. The node is required to adopt miniaturization design and use battery to supply power. In order to reduce the energy consumption, improve the energy efficiency and prolong the lifetime of nodes and networks, it is very difficult to replenish power after deployment.
The energy consumption of WSN can be divided into "necessary energy consumption" and "unnecessary energy consumption". Essential energy consumption has three purposes: (i) sending and receiving data; (ii) processing query requests; (iii) forwarding queries or packets to neighboring nodes. (iii) interception; (iv) generation or processing of control packets. There are ways to save energy in networks by reducing "unnecessary energy consumption" or "necessary energy consumption." These energy-saving methods can be divided into three categories: first, energy efficient routing technology; second, data processing technology; third, topology control technology. Both of them are effective ways to save energy and improve network energy efficiency, but there are still many theoretical barriers and technical bottlenecks to be solved.
In order to meet the challenges of WSN in energy-saving technology, a clustering routing protocol with balanced energy consumption is proposed based on the comprehensive analysis of WSN clustering routing principles and methods to prolong the network lifetime. In the application environment, the strategy and method of data fusion in wireless sensor networks are studied, and a high energy-efficient data fusion strategy based on clustering routing and a fuzzy comprehensive index data fusion algorithm based on coefficient of variation are proposed. The main contents and innovative achievements of the paper include:
(1) Clustering is an effective way to save energy and improve energy efficiency in wireless sensor networks. The traditional cluster head selection in clustering routing protocols is mostly realized by generating random numbers and comparing them with thresholds. This method leads to the randomness of cluster head selection, the number of cluster heads and the fluctuation of performance parameters in wireless sensor networks. Based on the equal clustering protocol, two clustering protocols, PNSCH and PCHIF, are proposed. PNSCH sends the residual energy information to the current cluster head in the form of report packets at the end of each round, and selects the node with the largest residual energy in the cluster as the next round. Cluster head. PCHSIF takes residual energy and distance from the base station as the basis of cluster head determination, and sends it to the cluster head of the current round for the next round of cluster head determination, and the number of cluster heads is adjusted to the optimal value. Although it consumes a certain amount of energy, it can achieve the goal of balancing the energy consumption of the network and saving the overall energy consumption of the network.
(2) Reducing the amount of data communication is also an effective way to increase energy consumption. The amount of data communication is closely related to the amount of information collected by sensors, which is redundant. Data fusion technology can solve the problem of data redundancy. According to the fusion strategy and algorithm, three data fusion strategies for wireless sensor networks (WSNs) in random deployment environment are proposed, and their performance in energy efficiency is analyzed. A data fusion algorithm for comprehensive evaluation of water quality in the whole monitoring area is proposed, which is based on fuzzy theory and adopts coefficient of variation method. The weights of different water quality indexes are determined, and a weighted comprehensive average algorithm is designed to weaken the influence of local specific data on the comprehensive evaluation of the whole monitoring area. Finally, the fuzzy comprehensive index is used to characterize the comprehensive evaluation category of local water quality and the comprehensive evaluation category of the whole region.
(3) The concept of energy efficiency evaluation for clustering routing protocols in wireless sensor networks is proposed for the first time, and the energy efficiency parameter system is studied. The multi-objective decision theory is introduced into the study of energy efficiency evaluation for clustering routing protocols in wireless sensor networks. Taking LEAC, LEACH-C, SEP and HEED clustering protocols as examples, the effectiveness of the energy efficiency comprehensive evaluation method is verified in three aspects: the selection of energy efficient clustering routing protocols under the same deployment conditions, the selection of energy efficient deployment schemes for the same clustering routing protocol, and the selection of energy efficient schemes for different clustering routing protocols under different deployment conditions. Choose.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5
本文编号：2190857
[Abstract]:Wireless Sensor Network (WSN), which is distributed randomly in the monitoring environment, works together and forms a network by means of wireless communication, and sends the effective information to the base station after simple processing such as calculation and fusion, micro-sensor, low-power electronic technology and radio frequency communication technology. The development of WSN promotes the emergence and vigorous development of WSN. Although WSN has a wide application prospect because of its low power consumption and flexible self-organization, the application environment of WSN is mostly complex and changeable, so it is not suitable to deploy the engineering application and environmental monitoring of wired network. The node is required to adopt miniaturization design and use battery to supply power. In order to reduce the energy consumption, improve the energy efficiency and prolong the lifetime of nodes and networks, it is very difficult to replenish power after deployment.
The energy consumption of WSN can be divided into "necessary energy consumption" and "unnecessary energy consumption". Essential energy consumption has three purposes: (i) sending and receiving data; (ii) processing query requests; (iii) forwarding queries or packets to neighboring nodes. (iii) interception; (iv) generation or processing of control packets. There are ways to save energy in networks by reducing "unnecessary energy consumption" or "necessary energy consumption." These energy-saving methods can be divided into three categories: first, energy efficient routing technology; second, data processing technology; third, topology control technology. Both of them are effective ways to save energy and improve network energy efficiency, but there are still many theoretical barriers and technical bottlenecks to be solved.
In order to meet the challenges of WSN in energy-saving technology, a clustering routing protocol with balanced energy consumption is proposed based on the comprehensive analysis of WSN clustering routing principles and methods to prolong the network lifetime. In the application environment, the strategy and method of data fusion in wireless sensor networks are studied, and a high energy-efficient data fusion strategy based on clustering routing and a fuzzy comprehensive index data fusion algorithm based on coefficient of variation are proposed. The main contents and innovative achievements of the paper include:
(1) Clustering is an effective way to save energy and improve energy efficiency in wireless sensor networks. The traditional cluster head selection in clustering routing protocols is mostly realized by generating random numbers and comparing them with thresholds. This method leads to the randomness of cluster head selection, the number of cluster heads and the fluctuation of performance parameters in wireless sensor networks. Based on the equal clustering protocol, two clustering protocols, PNSCH and PCHIF, are proposed. PNSCH sends the residual energy information to the current cluster head in the form of report packets at the end of each round, and selects the node with the largest residual energy in the cluster as the next round. Cluster head. PCHSIF takes residual energy and distance from the base station as the basis of cluster head determination, and sends it to the cluster head of the current round for the next round of cluster head determination, and the number of cluster heads is adjusted to the optimal value. Although it consumes a certain amount of energy, it can achieve the goal of balancing the energy consumption of the network and saving the overall energy consumption of the network.
(2) Reducing the amount of data communication is also an effective way to increase energy consumption. The amount of data communication is closely related to the amount of information collected by sensors, which is redundant. Data fusion technology can solve the problem of data redundancy. According to the fusion strategy and algorithm, three data fusion strategies for wireless sensor networks (WSNs) in random deployment environment are proposed, and their performance in energy efficiency is analyzed. A data fusion algorithm for comprehensive evaluation of water quality in the whole monitoring area is proposed, which is based on fuzzy theory and adopts coefficient of variation method. The weights of different water quality indexes are determined, and a weighted comprehensive average algorithm is designed to weaken the influence of local specific data on the comprehensive evaluation of the whole monitoring area. Finally, the fuzzy comprehensive index is used to characterize the comprehensive evaluation category of local water quality and the comprehensive evaluation category of the whole region.
(3) The concept of energy efficiency evaluation for clustering routing protocols in wireless sensor networks is proposed for the first time, and the energy efficiency parameter system is studied. The multi-objective decision theory is introduced into the study of energy efficiency evaluation for clustering routing protocols in wireless sensor networks. Taking LEAC, LEACH-C, SEP and HEED clustering protocols as examples, the effectiveness of the energy efficiency comprehensive evaluation method is verified in three aspects: the selection of energy efficient clustering routing protocols under the same deployment conditions, the selection of energy efficient deployment schemes for the same clustering routing protocol, and the selection of energy efficient schemes for different clustering routing protocols under different deployment conditions. Choose.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5
【参考文献】
相关期刊论文 前10条
1 吕云峰;马仁智;金宁;;一种用于水质监测的WSNs节点的设计与实现[J];传感器与微系统;2011年01期
2 梁焕焕;熊庆宇;石为人;;基于无线传感器网络的水质在线监测系统研究[J];传感器与微系统;2011年05期
3 周雅琴;谭定忠;;无线传感器网络应用及研究现状[J];传感器世界;2009年05期
4 何学文;郑乐平;孙汗;;无线传感器网络簇内多传感器数据融合算法[J];传感器与微系统;2014年01期
5 丰景春,陈立民,胡肇枢;水利工程评标综合评分法及其模型研究[J];河海大学学报(自然科学版);2003年04期
6 廖鹰;齐欢;王晓红;李伟群;;基于距离和分布的无线传感器网络分簇算法[J];华中科技大学学报(自然科学版);2012年06期
7 林志贵,徐立中,黄凤辰,王鸿彦;基于D-S理论的多源水质监测数据融合处理[J];计算机工程与应用;2004年10期
8 蒋畅江;石为人;向敏;唐贤伦;;基于PSO的无线传感器网络节能分簇协议[J];计算机工程;2010年08期
9 彭静;刘光祜;谢世欢;;无线传感器网络路由协议研究现状与趋势[J];计算机应用研究;2007年02期
10 黄建清;王卫星;姜晟;孙道宗;欧国成;卢康榉;;基于无线传感器网络的水产养殖水质监测系统开发与试验[J];农业工程学报;2013年04期
相关博士学位论文 前8条
1 姜延吉;多传感器数据融合关键技术研究[D];哈尔滨工程大学;2010年
2 魏博;无线传感器网络节能机制关键技术研究[D];解放军信息工程大学;2009年
3 孙亭;基于节能的无线传感器网络路由机制及算法研究[D];哈尔滨工程大学;2007年
4 张振江;关于无线传感器网络节能的几个关键问题研究[D];北京交通大学;2008年
5 马震;关于无线传感器网络节能的若干关键问题研究[D];北京交通大学;2009年
6 孙大洋;无线传感器网络节能问题研究[D];吉林大学;2010年
7 罗俊;无线传感器网络性能建模与优化设计[D];上海交通大学;2010年
8 周新莲;基于分簇技术的移动无线传感器网络数据收集协议研究[D];中南大学;2010年
，本文编号：2190857
本文链接：https://www.wllwen.com/kejilunwen/wltx/2190857.html
