基于三角划分的复杂3D山体表面定位算法  

Triangulation-based localization algorithm over complex 3D terrains

作　　者：王瑞锦[1] 秦志光[1] 包红来[1] 王聪[1] 

机构地区：[1]电子科技大学计算机科学与工程学院,成都611731

出　　处：《计算机应用研究》2013年第9期2823-2826,2830,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61133016);国家重大专项资助项目(2011ZX03002-002-03)

摘　　要：在复杂环境下,网络呈现出三维分布、链路非确定、应用场景异构、部署区域非结构化等新特点,传统的定位技术已无法适用。不规则的网络形状很可能导致不准确的传感器节点定位结果,从而导致定位误差较大。提出一种基于三角划分和MDS的3D节点定位算法。首先,基于三角划分的方法将不规则的3D网络模型分解为若干个较规则的三角区域;其次,在每一个三角区域上采用改进的MDS算法对节点进行定位,从而建立起局部相对位置地图;最后,合并三角子区域,建立起整个网络全局的位置地图。实验结果表明,该算法有效地提高了定位精度,具有一定的实用意义。In a.complex environment, the network appears as the situation of three-dimensional distribution, on-deterministic li^k, heterogeneous application scenarios, the unstructured deployment region and other new features. Thus, traditional posi- tioning technology has been unable to apply. The irregular shape of the network is likely to lead to inaccurate sensor node lo- calization results, causing larger localization errors. This paper proposed a 3D localization algorithm based on the triangulation model and MDS. First, it decomposed 3D network model into a number of regular triangle sub-areas by the triangulation mo- del, then used the MDS algorithm in each sub-area to obtain every node＇ s location, and could establish a local relative posi- tion of the map. Finally, it merged the sub-area to establish the location map of the entire global network. The experimental results show that the alzorithm is effective to improve the localization accuracy, and has somepractical applications.

关 键 词：无线传感器网络 三维定位 多维尺度 全局定位 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]