应用三维弹簧系统模型的锚节点部署优化

为使随机部署的三维无线传感器网络中锚节点的分布更加合理,提高未知节点定位精度,针对锚节点部署进行优化。通过构建弹簧系统模型,将锚节点抽象为通过弹簧相连接的点,使部分锚节点在合力作用下进行伸缩运动,达到提高网络性能的目的。当锚节点部署优化完成后,应用近似三角形内点测试( APIT)和DV-HOP( Distance Vector-hop)算法测试优化前后的节点定位精度。仿真结果表明,三维空间下的锚节点经过弹簧系统模型的部署优化后,锚节点网络覆盖率和定位覆盖率均得到了提高,网络平均连通度有所提升,且定位精度显著提高。     

WSN节点定位算法改进研究

在无线传感网络(WSN)中,节点的定位是一个非常之关键的问题,而DV-Hop算法是节点定位诸多算法中最主要的一个算法,但它计算未知节点到锚节点的距离时存在些许问题,所以提出了一个新的计算节点间距离的方法,考虑到在未知节点到锚节点路径上三个连续的节点所形成的角度的影响,使得距离的计算更精确.通过仿真可以得知这个改进算法能有效地提高未知节点定位的覆盖率和精度.     

一种新型三维传感器网络质心定位算法

现有的二维质心算法不能有效应用于三维无线传感器网络,针对这一问题,提出了基于虚拟节点的三维质心定位算法(Virtual centroid-3D).该算法将二维质心算法和三维网络模型相结合,在此基础上求出包含未知节点的四面体并引入虚拟节点的概念.运用这一方法,在设置相同的锚节点前提下,可以提供更多的已知节点,从而提高节点的定位精度.仿真结果表明该算法可显著减小定位误差,并能够高效地适应节点分布不均的三维网络环境.     

一种基于可移动锚节点的DV-HOP改进定位算法

当节点不均匀分布时,DV-Hop的定位精度较差。针对DV-Hop定位算法的缺陷,提出一种基于移动锚节点的改进DV-Hop定位算法。在网络中引入具有一定移动能力的锚节点,并构建锚节点之间的虚拟力模型,锚节点受到虚拟力作用发生移动,从而均匀的分布于整个网络,修正了DV-Hop对不均与分布网络适应性差的特点。仿真实验表明,与原始算法相比改进后的算法定位精度有较大提高。     

基于加权最小二乘变尺度法的无线传感器网络定位

为了降低整个无线传感器网络的成本和减小测量误差对定位精度的影响,一般传统做法是把已定位的未知节点升级为信标节点,再对其他节点进行定位,但此情况会造成累积误差。为了减少累积误差,提出了一种加权的最小二乘变尺度定位算法,该算法首先利用加权最小二乘法对未知节点进行位置估计,然后把定位的未知节点升级为信标节点,再对剩下的未知节点进行位置估计,最后利用拟牛顿法对估计出来的位置进行优化。仿真结果表明,该算法能有效地减少测距误差和累积误差,降低网络成本,提高网络覆盖率和传感节点的定位精度,并且该算法不增加额外硬件设备,易于实现。     

基于几何学的无线传感器网络定位算法

提出一种基于几何学的无线传感器网络(WSN)定位算法。把网络区域中的节点分为锚节点和未知节点,假设在定位空间中有n个锚节点,由于受到几何学的限制,实际可行的锚节点序列是有限的,因此利用一种几何方法判断锚节点间的位置关系,从而选取最优的锚节点序列,能够更精确地确定未知节点的位置,并且分析了待定位节点的邻居锚节点数量对定位精度的影响。仿真结果表明,与已有的APS(Ad-Hoc positioning system)定位算法相比,该算法可有效地降低平均定位误差和提高定位覆盖度。     

基于移动锚节点的无线传感网络节点定位算法

在无线传感网络定位算法中,锚节点位置决定了节点定位精度。为此,提出基于高斯-Markov模型的移动锚节点的节点定位(GM-MAL)算法。GM-MAL算法基于高斯-Markov移动模型,提出自适应锚节点的移动路径规划,通过速度调整策略、垂直平分线策略、虚斥力策略以及虚引力策略规划路径。在定位阶段,将非凸优化问题转化为双凸形式,再利用交替最小算法(AMA)求解,进而获取更短的锚节点移动路径。实验数据表明,引入虚引力策略提高了路径规划精度,覆盖了更多的监测区域。此外,相比于线性算法,GM-MAL的定位精度得到提高。     

无线传感器网络中基于锚节点功率调节的加权质心定位算法

提出了一种基于锚节点功率调节的加权质心定位算法,通过锚节点的功率调节确定各个锚节点对于未知节点的影响力因子,并将其作为权重计算未知节点的位置,体现了不同锚节点为未知节点位置计算结果的影响.仿真表明,该算法减小了节点的平均定位误差,是一种适合于无线传感器网络的定位方法.     

无线传感器网络中DV-Hop定位算法的改进

无线传感器网络中基于无需测距的节点定位算法定位精度不高,一般应用在粗精度定位中。为了提高基于无需测距的DV-Hop算法定位精度,利用最小均方差准则改进算法,通过修改指数值精化平均每一跳距离,提出不同通信半径、不同锚节点覆盖率下的最佳指数值概念,并应用在一种锚节点均匀分布环境中,进一步提高定位精度。MTLAB仿真结果表明,在最佳指数值下,改进的算法在不同锚节点覆盖率、不同通信半径下能提高定位精度,同时不会增加节点能量消耗与硬件成本。     

传感器网络的粒子群优化定位算法

无线传感器网络定位问题是一个基于不同距离或路径测量值的优化问题。由于传统的节点定位算法采用最小二乘法求解非线性方程组时很容易受到测距误差的影响,为了提高节点的定位精度,将粒子群优化算法引入到传感器网络定位中,提出了一种传感器网络的粒子群优化定位算法。该算法利用未知节点接收到的锚节点的距离信息,通过迭代方法搜索未知节点位置。仿真结果表明,该算法有效地抑制了测距误差累积对定位精度的影响,提高了节点的定位精度。     

Research on mobile strategy of anchor node based on weighted virtual force model

The existing mobility strategy of the anchor node in wireless sensor network (WSN) has the shortcomings of too long moving path and low positioning accuracy when the anchor node traverses the network voids area.A new mobility strategy of WSN anchor node was proposed based on an improved virtual forces model.The number of neighbor nodes and the distance between the neighbor nodes to the anchor nodes were introduced as their own dense weight attributes.The unknown nodes intensity was used as weights to improve the traditional virtual force model.Meantime the distance-measuring error ε was taken into account.The optimal distribution,direction selection,shift step length and fallback strategy of anchor node could be analyzed by the trilateration.Using the number of virtual beacon received by the unknown node and the distance between the unknown node to the anchor node calculate the virtual force.Then according to the virtual force,the direction was chosen and the anchor nodes were moved.Simulation experiments show that the strategy can make the anchor nodes move according to the specific circumstances of unknown node distribution.It has a high positioning accuracy and strong adaptability.It can successfully shorten the path of the anchor node movement and reduce the number of virtual beacon.Moreover it can effectively avoid the anchor node to enter the network voids area and reduce the number of collinear virtual anchor nodes.     

无线传感器网络一种改进的convex定位算法

以convex(凸规划)定位算法为基础,针对range-free定位算法中anchor(已知节点)比例低带来的定位精度低、网络覆盖率低的问题,提出了二跳信息改进定位算法。该算法中,未知节点在通信中加入自身邻居anchor的ID和位置信息并发送给邻居节点,相应的邻居节点从中确定自己的二跳邻居anchor,并利用二跳邻居anchor的二跳通信范围来减小未知节点的可能存在区域,进而提高未知节点的定位精度。仿真表明,二跳信息改进定位算法在anchor节点比例较低情况下能有效提高定位精度,而在anchor节点比例较高时接近原convex算法定位精度,并且网络规模越大这种提高越显著。     

Moving anchor node localization algorithm based on network connectivity

In order to better solve the contradiction between precision of localization and the number of anchor nodes in wireless sensor network,a mobile anchor node localization technology based on connectivity was proposed.First,the coverage characteristic of the network nodes was analyzed,and a critical value was found between the mobile step and the anchor node communication radius,mobile anchor nodes＇ coverage characteristic would change when near this critical value.Second,a mobile anchor node followed a planning path to form a positioning area seamless coverage was used.Finally,when there was no need for high-precision technology,node position would been estimated according with the connectivity of the network and the receiving information of the node.The simulation results show that the proposed algorithm can realize coarse-grained localization,and paths perform complete localization.     

A fast handover protocol for 6LoWPAN wireless mobile sensor networks

Node localization is one of the most critical issues for wireless sensor networks, as many applications depend on the precise location of the sensor nodes. To attain precise location of nodes, an improved distance vector hop (IDV-Hop) algorithm using teaching learning based optimization (TLBO) has been proposed in this paper. In the proposed algorithm, hop sizes of the anchor nodes are modified by adding correction factor. The concept of collinearity is introduced to reduce location errors caused by anchor nodes which are collinear. For better positioning coverage, up-gradation of target nodes to assistant anchor nodes has been used in such a way that those target nodes are upgraded to assistant anchor nodes which have been localized in the first round of localization. For further improvement in localization accuracy, location of target nodes has been formulated as optimization problem and an efficient parameter free optimization technique viz. TLBO has been used. Simulation results show that the proposed algorithm is overall 47, 30 and 22% more accurate than DV-Hop, DV-Hop based on genetic algorithm (GADV-Hop) and IDV-Hop using particle swarm optimization algorithms respectively and achieves high positioning coverage with fast convergence.     

WSN中基于最佳指数的加权DV-Hop算法

DV-Hop算法是一种低成本、低定位精度的无需测距定位算法,在粗精度定位中应用广泛。为提高DV-Hop算法定位精度,从减小锚节点的平均每一跳距离误差和减小未知节点平均每一跳校正值误差两方面考虑。首先,用最佳指数值下的公式计算锚节点平均每一跳距离。然后,将未知节点的校正值加权处理,使所有的锚节点根据与未知节点距离的远近影响校正值的大小。MATLAB实验证明,改进的基于最佳指数值下的加权DV-Hop算法比DV-Hop算法、加权DV-Hop、最佳指数值下DV-Hop算法定位精度分别提高2%左右、1.65%左右、1.15%左右,同时不会增加网络硬件成本。