无线传感器网络中基于网格法与椭圆测量模型的无源感知定位算法研究

发布时间：2018-04-28 12:41

本文选题：无线传感器网络 + RSSI　；参考：《安徽大学》2017年硕士论文

【摘要】：随着无线传感器网络技术的不断发展,无线传感器网络越来越多地应用于环境感知、目标监测等方面。已有的无线传感器定位技术多为有源定位,定位目标需要绑定有源器件或电子标签,在条件受限的环境中适用性不高。近年来利用无线传感器网络节点的接收信号强度指示RSSI(Received Signal Strength Indication)进行无源感知定位(Device-free Localization)的概念被提出:通过目标对无线信号的吸收、反射、折射作用而造成的链路RSSI变化,实现对感知目标位置的估计。无源感知定位使目标脱离了有源电子器件的限制,具有更强的灵活性与适用性,其中定位算法成为无源感知定位研究的重点。本文首先对无线传感器网络与无源感知定位技术进行了概述,对各类定位算法模型进行了分类与比较,重点研究了基于RSSI的无源感知定位算法,将网格法(Sensor Grid Array)与椭圆测量模型(Ellipse Measurement Model)结合,提出了基于网格法与椭圆测量模型的无源感知定位算法。算法首先根据网格法定位原理,利用传感器节点链路对感知区域进行划分,再通过比较链路RSSI的相对变化对链路进行筛选并结合椭圆测量模型绘制圆环线,依据椭圆交点确定感知目标位置。在确定算法的基本流程后,需对所使用的椭圆模型参数进行调整以适应算法的整体过程。因此本文在以CC2530芯片为核心的ZigBee无线传感器网络实验平台上进行定位实验,通过对定位误差数据的收集与分析,实现了模型参数的反向推导,确定了算法的最终形式。算法确定后,在原实验平台的基础上分别改变传感器节点的数目与感知区域面积,重新进行无源感知定位实验,完成实验数据收集并配合MATLAB进行算法仿真与模拟定位,验证了算法的有效性。通过定位误差分析与算法比较可以发现:基于网格法与椭圆测量模型的无源感知定位算法相较于传统的网格法能够以较低的传感器节点密度获得相近的定位精度,但会使定位产生边缘效应;相较于扫描成像法,基于网格法与椭圆测量模型的无源感知定位算法能够以更低的节点硬件成本获得更高的定位精度。
[Abstract]:With the development of wireless sensor network technology, wireless sensor network is more and more used in environmental perception, target monitoring and so on. Most of the existing wireless sensor localization techniques are active localization, the target location needs to bind active devices or electronic tags, so it is not suitable for the environment with limited conditions. In recent years, using the received signal strength of wireless sensor network nodes to indicate the RSSI(Received Signal Strength Indication) for passive sensing positioning has been proposed: the link RSSI changes caused by the target's absorption, reflection and refraction of the wireless signal. The estimation of target location is realized. Passive sensing localization makes the target get rid of the limitation of active electronic devices, so it has more flexibility and applicability, among which the localization algorithm becomes the focus of passive sensing localization research. Firstly, this paper summarizes the wireless sensor network and passive sensing localization technology, classifies and compares all kinds of localization algorithm models, and focuses on the passive sensing location algorithm based on RSSI. A passive sensing location algorithm based on mesh method and elliptic measurement model is proposed by combining the mesh method and Ellipse Measurement Model. Firstly, according to the principle of grid positioning, the sensor node link is used to divide the sensing region, then the link is screened by comparing the relative changes of link RSSI and the circle line is drawn by combining with the elliptical measurement model. The position of the perceived target is determined according to the intersection of the ellipse. After determining the basic flow of the algorithm, the parameters of the elliptic model should be adjusted to fit the whole process of the algorithm. Therefore, this paper carries on the localization experiment on the ZigBee wireless sensor network experiment platform with CC2530 chip as the core. Through collecting and analyzing the positioning error data, the inverse derivation of the model parameters is realized, and the final form of the algorithm is determined. After the algorithm is determined, on the basis of the original experimental platform, the number of sensor nodes and the area of sensing area are changed, and the passive sensing localization experiment is carried out again, and the experimental data are collected and the algorithm is simulated and simulated with MATLAB. The validity of the algorithm is verified. Through the analysis of the location error and the comparison of the algorithm, it can be found that compared with the traditional mesh method, the passive sensing localization algorithm based on mesh method and elliptical measurement model can obtain similar positioning accuracy with lower sensor node density. Compared with the scanning imaging method, the passive sensing localization algorithm based on mesh method and elliptical measurement model can achieve higher localization accuracy with lower node hardware cost.
【学位授予单位】：安徽大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN929.5

【参考文献】