基于RSSI的WSN定位算法及测距影响因素的研究

发布时间：2018-01-16 08:18

本文关键词：基于RSSI的WSN定位算法及测距影响因素的研究　出处：《吉林大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着物联网概念的普及和研究的深入，作为其核心技术之一的无线传感器网络技术更加引起各国研究人员的关注热情，在军事、医疗、建筑和交通等领域的应用层出不穷，这些应用的共同特点是都需要获取事件发生的位置信息，这就涉及到无线传感器网络定位技术。一般来说，WSN的定位可以分为两种类型：基于距离的和与距离无关的。与距离无关的定位方法不需要测距，这降低了硬件成本，但它的定位精度比基于距离的定位方法要差一些。基于RSSI测距的定位方法的优势在于其便捷和低成本的特性，如果能采取一定的措施来降低测距和定位算法带来的误差，这种定位方法将具有更加广阔的商业应用前景。本文的具体研究内容和创新工作如下： 1.系统论述了WSN的概念、体系结构、特点和在不同领域中的比较有特点的应用。指出无线传感器网络定位技术的研究意义，详细阐述了WSN发展过程中诞生的种种定位系统和定位算法，总结国内外最新的研究现状。 2.研究了无线电传播的几类数学模型，做实验收集数据拟合出RSSI测距模型。探究了影响RSSI测距的两个因素：天线的方向和人移动的速度。通过做实验收集数据，统计出RSSI的平均值和方差，利用上述信息和RSSI的波动范围来确定不同程度的因素对RSSI值的影响程度。 3.从分析基于RSSI的加权质心定位算法入手，针对可能引起定位误差的环节，提出了三处改进：对大量的RSSI值进行常态分布处理，划定高概率发生区域之后再求均值；对RSSI测距模型得出的距离引入差分修正模型，用参考节点的测距误差因子来修正未知节点的测距；对三边定位法定位出来的多个未知节点的近似坐标，综合考虑三边定位时影响定位精度的因素，最后选择三种权值因子进行加权：距离权值、角度权值、面积和边长权值，再对计算出来的三个坐标进行融合，，求得未知节点的最终估计位置。 4.通过Matlab平台对改进算法进行了仿真，对比基于RSSI的加权质心定位算法和基于RSSI的差分修正质心定位算法在定位误差方面的表现，并分别在不同的信标节点个数的情况下进行了比较。另外还对比了改进算法在最后一步坐标融合时采用不同权值系数组合的定位误差，确定了较优的组合方案。最终通过仿真实验验证了改进算法在减小定位误差上的优势。
[Abstract]:With the popularity of the concept of the Internet of things and the depth of research, as one of its core technologies, wireless sensor network technology has attracted more and more attention of researchers from all over the world, in the military, medical care. There are many applications in the field of architecture and transportation. The common feature of these applications is that they need to obtain the location information of events, which involves the location technology of wireless sensor networks. The location of WSN can be divided into two types: distance-based and range-independent. Range-independent localization methods do not require ranging, which reduces the cost of hardware. However, the positioning accuracy is lower than that based on distance. The advantage of the location method based on RSSI is its convenience and low cost. If some measures can be taken to reduce the errors caused by ranging and localization algorithms, this method will have a wider commercial application prospect. The specific research contents and innovative work of this paper are as follows: 1. The concept, architecture, characteristics and application of WSN in different fields are discussed systematically. The research significance of wireless sensor network localization technology is pointed out. In this paper, various localization systems and localization algorithms born during the development of WSN are described in detail, and the latest research status at home and abroad is summarized. 2. Several mathematical models of radio propagation are studied. The experimental data were collected to fit the RSSI ranging model. Two factors affecting the RSSI ranging were explored: the direction of the antenna and the speed of human movement. The average value and variance of RSSI are calculated, and the influence of different factors on RSSI value is determined by using the above information and the fluctuation range of RSSI. 3. Based on the analysis of weighted centroid localization algorithm based on RSSI, three improvements are put forward: dealing with the normal distribution of a large number of RSSI values. The mean value is obtained after the high probability area is delineated; A differential correction model is introduced to the distance obtained from the RSSI ranging model, and the ranging error factor of the reference node is used to correct the distance measurement of the unknown node. For the approximate coordinates of many unknown nodes located by the trilateral localization method, the factors that affect the positioning accuracy are considered synthetically. Finally, three weight factors are selected for weighting: distance weight value, angle weight value. The area and the weight of the side length are fused to the calculated three coordinates and the final estimated position of the unknown node is obtained. 4. The improved algorithm is simulated on Matlab platform, and the performance of weighted centroid location algorithm based on RSSI and differential modified centroid location algorithm based on RSSI is compared. In addition, the location errors of the improved algorithm in the last step coordinate fusion using different weight coefficients combination are compared respectively in the case of different number of beacon nodes. Finally, the advantages of the improved algorithm in reducing the location error are verified by simulation experiments.
【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】：TN929.5;TP212.9

【参考文献】