约束条件下传感器节点覆盖和部署问题研究
发布时间：2018-08-12 14:25
【摘要】：近年来,无线传感器网络(Wireless Sensor Networks,WSNs)发展迅速,并逐渐成为连接物理世界和数字世界的一个桥梁。节点部署问题是WSNs设计的第一步,它对WSNs的覆盖,连通,能耗和生命周期等有着很大的影响。覆盖是WSNs研究的基本问题之一,它严重影响着网络的能耗与生存周期。WSNs的节点部署问题就是要寻找一个满足目标条件的最佳拓扑结构,一个好的拓扑结构不但能够提高WSNs的覆盖和连通性能,而且还能降低成本和提高网络的生命周期。在WSNs执行监测任务时,选择尽量少的工作节点,可以降低网络能耗和传感器节点间感知数据的冗余度。本文主要研究了保持部分连通的最少传感器节点覆盖问题和最少传感器节点部署问题。本文针对WSNs的节点覆盖和部署问题主要完成了如下工作:对于最少传感器节点部署问题,本文提出了一种公交车载网络(Bus-based Ad hoc Networks,BANETs)和无线传感器网络组成的混合网络模型,并在此网络模型上提出了改进的有边界的基于条状的传感器节点部署算法(MSSDB),MSSDB考虑了目标区域的边界问题,并针对传感器节点的感知半径和通信半径关系的不同,在基于条状和基于三角形部署方式之间切换。本文还分析了基站可以在一定的时间延迟内接收完全城的数据包。仿真实验表明:MSSDB比其他传统的传感器节点部署机制节省传感器节点,并且能够满足一定的时延限制。对于保持部分连通的最少传感器节点覆盖问题,本文提出了一种基于Connect Road Gain节点选择算法(SSCG)。通过把区域覆盖问题转化为目标点覆盖问题,贪婪式的选择能够覆盖最多目标点的传感器节点加入到结果集合中,直到结果集合中的传感器节点能够覆盖整个目标区域。然后判断结果集合中的每一个传感器节点是否和道路连通,如果不是,则根据节点的Connect Road Gain值大小,选择ConnectRoad Gain值最大的节点添加到结果集合,最终求得满足覆盖和部分连通的最少传感器节点集合。仿真实验表明:当通信半径不大于2倍的感知半径时,本文所提出的算法SSCG优于传统的保持连通和覆盖的节点选择算法(CBA);当通信半径远大于感知半径时,和CBA相比,SSCG没有优势。当网络中节点的密度比较大时,SSCG的优势比较明显,但是如果网络很稀疏,那么SSCG和CBA的性能都不好。
[Abstract]:In recent years, wireless sensor networks (Wireless Sensor Networks WSNs) have developed rapidly and become a bridge between the physical world and the digital world. Node deployment is the first step in the design of WSNs. It has great influence on the coverage, connectivity, energy consumption and life cycle of WSNs. Coverage is one of the basic problems in WSNs research. It seriously affects the network energy consumption and lifetime. WSNs node deployment problem is to find an optimal topology to meet the target conditions. A good topology can not only improve the coverage and connectivity of WSNs, but also reduce the cost and improve the lifetime of the network. When WSNs performs monitoring tasks, selecting as few working nodes as possible can reduce the network energy consumption and the redundancy of sensor nodes' perceptual data. In this paper, the problem of maintaining partially connected least sensor node coverage and least sensor node deployment is studied. The main work of this paper is as follows: for the problem of minimum sensor node deployment, a hybrid network model composed of Bus-based Ad hoc Networks (Bus-based) and Wireless Sensor Networks (WSN) is proposed. Based on this network model, an improved striped sensor node deployment algorithm, (MSSDB) / MSSDB, is proposed, which takes into account the boundary problem of the target region, and aims at the difference of the relationship between the sensor node's perceptual radius and the communication radius. Switch between stripe-based and triangular-based deployment. This paper also analyzes that the base station can receive the complete city data packet within a certain time delay. Simulation results show that: MSSDB can save sensor nodes and meet certain delay limits compared with other traditional sensor node deployment mechanisms. In this paper, a node selection algorithm based on Connect Road Gain (SSCG).) is proposed to cover the least sensor nodes with partially connected nodes. By transforming the region coverage problem into the target point coverage problem, the greedy selection of sensor nodes covering the most target points can be added to the result set until the sensor nodes in the result set can cover the entire target region. Then determine whether each sensor node in the result set is connected to the road, and if not, select the node with the largest ConnectRoad Gain value to add to the result set, depending on the size of the node's Connect Road Gain value. Finally, the set of minimal sensor nodes satisfying coverage and partial connectivity is obtained. Simulation results show that when the communication radius is less than 2 times the perceptual radius, the proposed algorithm SSCG is superior to the traditional node selection algorithm (CBA);), which maintains connectivity and coverage. When the communication radius is much larger than the perceptual radius, the proposed algorithm has no advantage over CBA. The advantage of SSCG is obvious when the density of nodes in the network is high, but if the network is sparse, the performance of SSCG and CBA is not good.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP212.9;TN929.5
本文编号：2179328
[Abstract]:In recent years, wireless sensor networks (Wireless Sensor Networks WSNs) have developed rapidly and become a bridge between the physical world and the digital world. Node deployment is the first step in the design of WSNs. It has great influence on the coverage, connectivity, energy consumption and life cycle of WSNs. Coverage is one of the basic problems in WSNs research. It seriously affects the network energy consumption and lifetime. WSNs node deployment problem is to find an optimal topology to meet the target conditions. A good topology can not only improve the coverage and connectivity of WSNs, but also reduce the cost and improve the lifetime of the network. When WSNs performs monitoring tasks, selecting as few working nodes as possible can reduce the network energy consumption and the redundancy of sensor nodes' perceptual data. In this paper, the problem of maintaining partially connected least sensor node coverage and least sensor node deployment is studied. The main work of this paper is as follows: for the problem of minimum sensor node deployment, a hybrid network model composed of Bus-based Ad hoc Networks (Bus-based) and Wireless Sensor Networks (WSN) is proposed. Based on this network model, an improved striped sensor node deployment algorithm, (MSSDB) / MSSDB, is proposed, which takes into account the boundary problem of the target region, and aims at the difference of the relationship between the sensor node's perceptual radius and the communication radius. Switch between stripe-based and triangular-based deployment. This paper also analyzes that the base station can receive the complete city data packet within a certain time delay. Simulation results show that: MSSDB can save sensor nodes and meet certain delay limits compared with other traditional sensor node deployment mechanisms. In this paper, a node selection algorithm based on Connect Road Gain (SSCG).) is proposed to cover the least sensor nodes with partially connected nodes. By transforming the region coverage problem into the target point coverage problem, the greedy selection of sensor nodes covering the most target points can be added to the result set until the sensor nodes in the result set can cover the entire target region. Then determine whether each sensor node in the result set is connected to the road, and if not, select the node with the largest ConnectRoad Gain value to add to the result set, depending on the size of the node's Connect Road Gain value. Finally, the set of minimal sensor nodes satisfying coverage and partial connectivity is obtained. Simulation results show that when the communication radius is less than 2 times the perceptual radius, the proposed algorithm SSCG is superior to the traditional node selection algorithm (CBA);), which maintains connectivity and coverage. When the communication radius is much larger than the perceptual radius, the proposed algorithm has no advantage over CBA. The advantage of SSCG is obvious when the density of nodes in the network is high, but if the network is sparse, the performance of SSCG and CBA is not good.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP212.9;TN929.5
【参考文献】
相关期刊论文 前1条
1 匡林爱;蔡自兴;;基于遗传算法的无线传感器网络重新部署方法[J];控制与决策;2010年09期
，本文编号：2179328
本文链接：https://www.wllwen.com/kejilunwen/wltx/2179328.html
