WSNs中基于无人机的强健节点定位算法

通过移动无人机(UAV)收集无线传感网络数据的方案已受到广泛关注,将感测的数据与产生此数据的传感节点位置关联起来是十分必要的。为此提出了基于无人机的强健节点定位算法(UAV-NL)。UAV-NL算法将UAV位置作为未知信息。传感节点接收由UAV在随机位置传输的beacon包,并记录接收信号强度指示(RSSI)矢量;通过理论推导2个RSSI矢量的范数距离与这2节点距离的线性关系;最后,通过RSSI值测距,并利用半定规划(SDP)算法估计节点位置。仿真结果表明,提出的UAV-NL算法即使在噪声信道条件下仍具有高的定位精确度。     

一种新的无线传感网络三维定位方法

定位是无线传感器网络的基础问题之一,文章提出利用均值法对接收信号强度指示(RSSI)数据进行处理,筛选出RSSI值较优的锚节点,以解决RSSI易受干扰的问题,减小RSSI的测距误差。在此基础上,提出动态修正三维三边测量方法。该方法利用筛选出的RSSI值较优的3个锚节点进行测距,在一个移动锚节点辅助下进行三维三边定位,提高定位精确度。仿真结果表明,与传统三边测量定位算法相比,此方法可明显减少定位误差。     

基于RSSI高斯滤波的LSSVR无线传感网络定位算法

为了降低基于接收信号强度指示(RSSI)测距误差对节点定位的影响,解决RSSI测距定位误差较大的问题,提出基于RSSI高斯滤波的最小二乘支持向量回归机LSSVR定位算法(LSSVR-GF-RSSI)。LSSVR-GF-RSSI算法先利用高斯函数滤除误差较大的RSSI值,筛选出较准确的RSSI值,再依据这些值计算未知节点离锚节点间的距离。将这些距离作为LSSVR的输入,建立基于RSSI测距的LSSVR定位算法模型,最终,估计未知节点的位置。仿真结果表明,提出的LSSVR-GF-RSSI算法能够有效地降低均方定位误差,比传统的基于RSSI的LSSVR定位算法减少了约12%~20%。     

基于BP神经网络和泰勒级数的室内定位算法研究

 在研究分析室内无线信号传播特性和传统的室内定位算法的基础上,提出了用BP神经网络来拟合室内无线信号传播模型,避免了对无线信号传播模型中参数A和n的不精确估计.在训练完成的BP神经网络的输入层输入接收信号强度值RSSI(Received Signal Strength Indicator),在输出层即可得到对应的距离值,再利用泰勒级数展开法确定盲节点的坐标位置.最终通过Matlab仿真和ZigBee平台实验验证了算法的可行性和有效性.     

基于干涉效应的无线传感器节点定位方法

针对无线传感器网络中节点定位的需求,提出了一种基于干涉效应的传感器节点定位算法。该方法仅需要利用节点的硬件设备,不需要增加额外的测量设备,各个节点独立进行自身位置估计,实现分布式定位。首先在传感器网络中设置若干个位置已知的网关锚节点和参考接收节点,不同位置两个网关锚节点在某个时隙内发射单音信号,就会在整个定位区域内形成一个干涉效应场。而各个待定位节点通过测量接收的到的干涉信号振动周期数就可以得到自身与各个网关节点距离差。通过多组锚节点分时隙发射单音信号,并配合参考接收节点,通过双曲线定位算法得到自身的坐标。通过对算法进行仿真分析,验证了算法的可行性。     

WSNs中基于通用回归神经网络的目标跟踪算法

基于接收信号强度指示(RSSI)的移动目标定位和跟踪常采用三边或角度测量定位技术。尽管该技术简单,易实施,但由于RSSI值与距离间的非线性关系,它们容易导致较大的定位误差。通用回归神经网络(GRNN)能够快速训练稀疏数据集。提出基于GRNN的移动目标跟踪(GMTT)算法,该算法依据GRNN处理RSSI与目标位置间的非线性关系,利用卡尔曼滤波(KF)修正目标位置。仿真实验结果表明,相比于RSSI+KF,GMTT算法可以有效地降低目标定位的根均方误差。     

多功率移动锚节点辅助的分布式节点定位方法

提出了一种移动锚节点辅助的分布式定位算法.与以前的基于移动锚节点的定位算法不同,此算法不需要任何测距技术支持.它是利用移动锚节点的功率控制,即以不同的发射功率发射信标信号,接收到信标信号的未知节点将这些信标信息转化为一系列二次不等式约束,然后通过凸优化技术求解这些不等式组来逼近未知节点位置的最佳估计.仿真结果表明,提出的距离无关的定位算法可适合实际定位情况且具有较高的定位精度.     

基于RSSI的Wi-Fi室内定位常用算法对比

WiFi定位是室内定位的方法之一,其观测值主要为WiFi的信号强度(received signal strength indication,RSSI),利用信号强度进行室内定位的方法基本上可以分为两种,一种根据模型计算接收节点与发射节点的距离,估计接收节点的位置;另一种是根据事先采集的信号强度指纹进行空间匹配。通过在不同场景下对两种定位算法的比较,基于信号强度指纹匹配的定位精度明显高于基于估计接收节点的定位精度,但与指纹匹配法相比,位置估计法更加灵活。     

基于临近锚节点修正的DBSCAN聚类加权定位算法

为了提高无线传感器网络节点的定位精确度,给出一种基于临近锚节点修正(CAAN)的具有噪声的基于密度的聚类(DBSCAN)加权定位算法.首先,在未知节点通信范围内的锚节点中选择三个构成三角形,证明当未知节点处在此三角形外接圆圆心位置时定位误差最小,然后据此选择合适的锚节点,结合滤波后的接收信号强度指示(RSSI)值进行定...     

一种改进型栅格划分的概率定位算法

基于接收信号强度指示(RSSI,Received Signal Strength Indicator)的定位算法,由于无线信号受到多径衰减、障碍物、噪声等因素的影响,定位误差值一直较大.通过对基于RSSI定位算法的分析,在矩形栅格划分概率定位算法的基础之上,提出一种改进型算法——三角形栅格划分概率定位算法.实验结果显示,节点定位误差进一步减小,进一步提高了节点定位精度.     

基于连通度和加权校正的移动节点定位算法

移动节点定位问题是无线传感器网络中的研究重点。针对移动节点定位误差大的问题,提出一种基于连通度和加权校正的移动节点定位算法。在未知节点移动过程中,根据节点间连通度大小选取参与定位的信标节点,利用加权校正方法修正RSSI测距信息,然后用最小二乘法对未知节点进行位置估计。仿真分析表明,节点通信半径和信标密度在一定范围内,该算法表现出良好的定位性能,定位精度明显提升。     

无线传感器网络差分修正定位算法的改进

提出了无线传感器网络中一种基于接收信号指示强度的改进差分修正算法,与传统的差分修正算法相比,在该算法中,通过各个信标节点分别作为差分参考节点进行定位,避免了单个差分参考节点对未知节点定位决定权过大。同时,提出加权因子的概念,体现了各差分参考点对定位效果的决定权。实验结果表明,改进的差分修正算法的定位精度和稳定性有明显提高。     

Implementation of an Efficient Artificial Bee Colony Algorithm for Node Localization in Unmanned Aerial Vehicle Assisted Wireless Sensor Networks

Node localization is a fundamental task in wireless sensor networks as it is useful for several localization based protocols and applications. Node localization using Global Poisoning System (GPS) employed fixed terrestrial anchor nodes suffers from high deployment cost and poor localization accuracy in GPS denied locations. These issues can be easily handled by deploying movable Unmanned Aerial Vehicles (UAVs). A movable UAV equipped with a single GPS module virtually increases number of anchor nodes and localizes a node at different locations. Hence, UAVs are cost effective and also provides high localization accuracy. As the flying altitude of UAV greatly influence localization accuracy, the present work firstly optimizes the flying height and then the node localization is defined as least square optimization problem using this optimal height. Since the classical received signal strength indicator based multilateration results high localization error, the least square localization using optimization techniques is found to be better alternative. The recently proposed Artificial Bee Colony (ABC) algorithm is a powerful optimization technique that can be applied for this optimization problem to achieve high accuracy. Thus, this paper aims at designing an ABC localization technique using UAV anchors to achieve minimum localization error. Further, we provide detailed simulation analysis to support the proposed ABC localization scheme.

     

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.     

Power efficient range-free localization algorithm for wireless sensor networks

Considering energy consumption, hardware requirements, and the need of high localization accuracy, we proposed a power efficient range-free localization algorithm for wireless sensor networks. In the proposed algorithm, anchor node communicates to unknown nodes only one time by which anchor nodes inform about their coordinates to unknown nodes. By calculating hop-size of anchor nodes at unknown nodes one complete communication between anchor node and unknown node is eliminated which drastically reduce the energy consumption of nodes. Further, unknown node refines estimated hop-size for better estimation of distance from the anchor nodes. Moreover, using average hop-size of anchor nodes, unknown node calculates distance from all anchor nodes. To reduce error propagation, involved in solving for location of unknown node, a new procedure is adopted. Further, unknown node upgrades its location by exploiting the obtained information in solving the system of equations. In mathematical analysis we prove that proposed algorithm has lesser propagation error than distance vector-hop (DV-Hop) and other considered improved DV-Hop algorithms. Simulation experiments show that our proposed algorithm has better localization performance, and is more computationally efficient than DV-Hop and other compared improved DV-Hop algorithms.     

基于改进RSSI测距的LSSVR三维WSN定位算法

针对基于RSSI测距的定位算法定位误差较大的问题,通过加入多组已知节点之间的距离和接收功率作为参考,提出了一种改进的RSSI测距算法,并将改进的RSSI测距作为最小二乘支持向量回归机LSSVR的输入向量,获得基于改进RSSI测距的LSSVR三维定位算法模型。MATLAB仿真结果表明,在节点随机分布的三维环境中,基于改进RSSI测距的LSSVR定位算法的定位误差比传统LSSVR定位算法减小了13.6%~21.2%,另外,可以通过增加已知节点数量等方法,进一步提高目标定位的准确性。     

无线传感器网络用于监测系统中的定位算法

从硬件设计入手,介绍了自主设计的以片上系统（SOC）STM32W108为核心的WSN节点,在此基础上采用了一种基于RSSI的加权质心定位算法实现了节点的自定位。该算法将RSSI测距和质心定位算法相结合,用测得的RSSI值作为质心定位的加权因子,合理体现了不同锚节点对定位未知节点的约束力。通过测试证明,该定位方法在较少的通信开销情况下具有较高的定位精度,且易于实现。