无线传感器网络中DV-Hop算法的改进研究
发布时间：2018-07-13 17:04
【摘要】：现在越来越多的人对无线传感器网络（Wireless Sensor Network，WSN）有了接触与了解。由最初的简单传感器到无线传感器现阶段无线传感器网络的发展，使得无线传感器网络在国际上被认为是继互联网之后的第二大网络。无线传感器网络与通信技术和计算机技术共同构成信息技术的三大支柱。由于通信技术、嵌入式技术及传感器网络的相互融合与使用，越来越多的行业将其作为研究新项目的手段依据。 在无线传感器网络中，最重要的功能是传感器节点自身的定位，需要网络能发现事件发生的具体位置。如果没有具体的位置信息，得到的数据信息也就没有意义。所以在无线传感器网络中，只有得到节点发布的自身位置信息和搜集到的数据才可以决定下一步做什么，才便于找到节点传输的路线轨迹。GPS全球定位系统已经发展的较为成熟，其定位准确率高、效率快、抗干扰能力强，但是由于GPS定位需要有一定的设施支持，能量消耗大，使用成本高，且GPS系统只适用于无遮挡的空旷环境，因此，研究节点的定位算法成了关键之处。无线传感器网络的布置环境及一些设备支持相对来说会简单些，对于定位采用设置的定位算法，与硬件设备的需求不会太紧密，这对于研究无线传感器网络的定位提供了很大的方便。 无线传感器网络的定位算法是近些年来研究的热点之一，根据网络的拓扑结构变化，节点自组织选择定位算法进行定位。技术人员需要从检测的目标节点获得关键数据信息，从信息中分析当前的节点处于环境中什么状态，这就需要目标节点将自身的位置反应给技术人员。例如森林大火，传感器节点不仅能发出火警信号，还需要将火灾的大致位置传给监测人员。因此对于定位算法，之前的研究人员、学者做出了很大的贡献，给出了多种定位算法。本文对常见的定位算法进行介绍，测距相关的及测距无关的分类方法中，针对典型的算法进行了详细的描述。最终将重点放在DV-Hop算法。 文章分析了DV-Hop算法误差产生的原因，就未知节点计算平均每跳距离估计值及未知节点定位计算提出了改进思想。其中，对于未知节点在利用周围锚节点计算距离时，选择一跳范围内的锚节点。原DV-Hop算法是未知节点接收第一个锚节点传来的平均每跳距离值，但第一个锚节点传来的值不一定是距离最近的。由于假设实验环境是一定的，节点是均匀分布的，，所以一跳范围内的锚节点传来的平均每跳距离值近似的可以看成是第一个锚节点传来的，算出平均的PJhopSize。这是与原算法的不同之处。其次用百分比是用来对锚节点求出的平均每跳距离进行修正的，得到Dhop。最后用PJhopSize和Dhop算加和平均得到最终的平均每跳距离PDhop。当未知节点用三边测量法或极大似然估计法算出估计坐标，往往会有很大的误差。针对误差过大影响定位精度的问题，需要找到一种算法能够快速找到并且能够优化最终的求解，从而达到使精度提高的目的。文章又提出了智能算法——基于差分进化的粒子群算法来对坐标结果进行优化。对于粒子群算法和差分进化粒子群算法，进行了详细的介绍。 文章在第五节就改进的算法PDDV-Hop进行了实验仿真验证。实验表明，改进的算法在定位精度上有一定的提高。但依旧存在一些不足之处，例如实验环境是假设节点散布均匀的情况下进行的，而现实生活中，节点的分布不会均匀。由于未知节点估算出来的坐标本身就存在很大的误差，智能算法就是改进修正也只是对误差值进行修正，减小误差范围，而不能更精确的接近目标位置。所以在之后的工作中，无论是对未知节点自身的定位还是用智能算法修正都有待进一步的研究。
[Abstract]:Nowadays, more and more people have contact and understanding of Wireless Sensor Network (WSN). From the initial simple sensor to the development of wireless sensor network at the present stage of wireless sensor, wireless sensor network is considered to be the second largest network after the Internet. Communication technology and computer technology constitute the three pillars of information technology. Because of the integration and use of communication technology, embedded technology and sensor networks, more and more industries use it as a means to study new projects.
In the wireless sensor network, the most important function is the location of the sensor node itself, the network can find the specific location of the event. If there is no specific location information, the data information is meaningless. So in the wireless sensor network, only the location information published by the node and the collection of the data are obtained. Data can decide what to do next, only to find the route path of node transmission,.GPS global positioning system has developed more mature, its positioning accuracy is high, efficiency is fast, anti-interference ability is strong, but because GPS positioning needs a certain facility support, high energy consumption, high cost, and GPS system is only applicable to no cover. Therefore, the location algorithm of the research node is the key point. The layout environment of the wireless sensor network and the support of some equipment are relatively simple. The location algorithm for positioning and the need of hardware devices will not be too close. This provides a great deal of research on the location of wireless sensor networks. It is convenient.
The location algorithm of wireless sensor networks is one of the hot topics in recent years. According to the changes in the topology of the network, the node is self-organizing and localize the location algorithm. The technical personnel need to obtain the key data information from the target node of the detection, and analyze the status of the current nodes in the environment from the information, which requires the goal. The node reacts its own position to the technical personnel. For example, the forest fire, the sensor nodes not only send out the fire alarm signal, but also need to transmit the rough location of the fire to the monitoring personnel. Therefore, the researchers and scholars have made a great contribution to the location algorithm, and the scholars have given a variety of location algorithms. This article introduces the typical algorithm in detail, and focuses on the DV-Hop algorithm.
In this paper, the reasons for the error of DV-Hop algorithm are analyzed. An improved idea is proposed for the estimated average per hop distance of unknown nodes and the location calculation of unknown nodes. The original DV-Hop algorithm is the first anchor node for the unknown node to choose the anchor node in the one hop range when the unknown node calculates the distance from the surrounding anchor nodes. The average per hop distance is transmitted, but the value of the first anchor node is not necessarily the nearest distance. Since the experimental environment is certain and the node is uniformly distributed, the average per jump distance derived from the anchor node within the one jump can be seen as the first anchor node, and the average PJhopSize. is calculated. The difference between the original algorithm and the original algorithm. Secondly, the percentage is used to correct the average per hop distance obtained by the anchor node. The Dhop. finally uses PJhopSize and Dhop to add and average the final average per hop distance PDhop.. When the unknown node uses the three edge measurement or maximum likelihood estimation to calculate the estimated coordinates, it is often very large. It is necessary to find an algorithm to quickly find and optimize the final solution in order to achieve the goal of improving the precision. In this paper, an intelligent algorithm, based on the differential evolution particle swarm optimization, is proposed to optimize the result of the sitting standard. The evolutional particle swarm optimization (PSO) is introduced in detail.
In the fifth section, the improved algorithm PDDV-Hop is verified by experimental simulation. The experiment shows that the improved algorithm has a certain improvement in the positioning accuracy. But there are still some shortcomings, for example, the experimental environment is assumed that the node is distributed uniformly, and the distribution of the nodes is not uniform in real life. There is a great error in the coordinates of the nodes estimated by the nodes. The intelligent algorithm is an improvement and correction, which only corrects the error value and reduces the error range, but can not be closer to the target position. Therefore, in the later work, it is necessary to further study the location of the unknown node itself or the correction of the intelligent algorithm. Study.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5
本文编号：2120118
[Abstract]:Nowadays, more and more people have contact and understanding of Wireless Sensor Network (WSN). From the initial simple sensor to the development of wireless sensor network at the present stage of wireless sensor, wireless sensor network is considered to be the second largest network after the Internet. Communication technology and computer technology constitute the three pillars of information technology. Because of the integration and use of communication technology, embedded technology and sensor networks, more and more industries use it as a means to study new projects.
In the wireless sensor network, the most important function is the location of the sensor node itself, the network can find the specific location of the event. If there is no specific location information, the data information is meaningless. So in the wireless sensor network, only the location information published by the node and the collection of the data are obtained. Data can decide what to do next, only to find the route path of node transmission,.GPS global positioning system has developed more mature, its positioning accuracy is high, efficiency is fast, anti-interference ability is strong, but because GPS positioning needs a certain facility support, high energy consumption, high cost, and GPS system is only applicable to no cover. Therefore, the location algorithm of the research node is the key point. The layout environment of the wireless sensor network and the support of some equipment are relatively simple. The location algorithm for positioning and the need of hardware devices will not be too close. This provides a great deal of research on the location of wireless sensor networks. It is convenient.
The location algorithm of wireless sensor networks is one of the hot topics in recent years. According to the changes in the topology of the network, the node is self-organizing and localize the location algorithm. The technical personnel need to obtain the key data information from the target node of the detection, and analyze the status of the current nodes in the environment from the information, which requires the goal. The node reacts its own position to the technical personnel. For example, the forest fire, the sensor nodes not only send out the fire alarm signal, but also need to transmit the rough location of the fire to the monitoring personnel. Therefore, the researchers and scholars have made a great contribution to the location algorithm, and the scholars have given a variety of location algorithms. This article introduces the typical algorithm in detail, and focuses on the DV-Hop algorithm.
In this paper, the reasons for the error of DV-Hop algorithm are analyzed. An improved idea is proposed for the estimated average per hop distance of unknown nodes and the location calculation of unknown nodes. The original DV-Hop algorithm is the first anchor node for the unknown node to choose the anchor node in the one hop range when the unknown node calculates the distance from the surrounding anchor nodes. The average per hop distance is transmitted, but the value of the first anchor node is not necessarily the nearest distance. Since the experimental environment is certain and the node is uniformly distributed, the average per jump distance derived from the anchor node within the one jump can be seen as the first anchor node, and the average PJhopSize. is calculated. The difference between the original algorithm and the original algorithm. Secondly, the percentage is used to correct the average per hop distance obtained by the anchor node. The Dhop. finally uses PJhopSize and Dhop to add and average the final average per hop distance PDhop.. When the unknown node uses the three edge measurement or maximum likelihood estimation to calculate the estimated coordinates, it is often very large. It is necessary to find an algorithm to quickly find and optimize the final solution in order to achieve the goal of improving the precision. In this paper, an intelligent algorithm, based on the differential evolution particle swarm optimization, is proposed to optimize the result of the sitting standard. The evolutional particle swarm optimization (PSO) is introduced in detail.
In the fifth section, the improved algorithm PDDV-Hop is verified by experimental simulation. The experiment shows that the improved algorithm has a certain improvement in the positioning accuracy. But there are still some shortcomings, for example, the experimental environment is assumed that the node is distributed uniformly, and the distribution of the nodes is not uniform in real life. There is a great error in the coordinates of the nodes estimated by the nodes. The intelligent algorithm is an improvement and correction, which only corrects the error value and reduces the error range, but can not be closer to the target position. Therefore, in the later work, it is necessary to further study the location of the unknown node itself or the correction of the intelligent algorithm. Study.
【学位授予单位】：吉林大学
【学位级别】：硕士
【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 杜晓明;陈岩;;无线传感器网络研究现状与应用[J];北京工商大学学报(自然科学版);2008年01期
2 杨凤;史浩山;朱灵波;赵洪钢;;一种基于测距的无线传感器网络智能定位算法[J];传感技术学报;2008年01期
3 石为人;贾传江;梁焕焕;;一种改进的无线传感器网络DV-Hop定位算法[J];传感技术学报;2011年01期
4 冯翔;陈国龙;郭文忠;;粒子群优化算法中加速因子的设置与试验分析[J];集美大学学报(自然科学版);2006年02期
5 王金鑫;赖旭芝;吴敏;;一种基于遗传算法的无线传感器网络定位新算法[J];计算技术与自动化;2007年04期
6 姚丹霖;端木庆敏;;无线传感器网络节点定位算法研究[J];计算机工程与科学;2008年03期
7 王驭风;王岩;;基于矢量和粒子群优化的传感器网络节点定位[J];计算机应用;2009年01期
8 邓彬伟;黄光明;;WSN中的质心定位算法研究[J];计算机应用与软件;2010年01期
9 万国峰;钟俊;杨成慧;;改进的RSSI测距和定位算法[J];计算机应用研究;2012年11期
10 刘明;包亚萍;刘汉义;;无线传感器网络中一种改进的DV-Hop算法[J];微计算机信息;2009年10期
本文编号：2120118
本文链接：https://www.wllwen.com/kejilunwen/wltx/2120118.html
