基于最小约束系统的分布式协作定位研究  

Research on Distributed Cooperative Localization Under Minimally Constraint System

作　　者：章平 王炎[2] 王桥[2] ZHANG Ping;WANG Yan;WANG Qiao(School of Computer and Information,Anhui Polytechnic University,Wuhu 241000,China;School of Information Science and Engineering,Southeast University,Nanjing 210096,China)

机构地区：[1]安徽工程大学计算机与信息学院,芜湖241000 [2]东南大学信息工程学院,南京210096

出　　处：《导航定位与授时》2019年第2期67-72,共6页Navigation Positioning and Timing

基　　金：国家科技重大专项基金(2016ZX03001022002);国家自然科学基金(61501005)

摘　　要：5G移动通信系统中,设备到设备的通信方式产生了节点间的信号测量,可用于提升无线定位的覆盖范围和定位精度。提出了一种基于最小约束系统的分布式协作定位方法,该方法首先利用已有算法估计节点参考位置,并据此构建最小约束确定网络的位置及朝向,然后在该约束下极大化由节点间测量引出的似然函数,获得位置估计。理论研究表明,该方法将由节点间测量确定的相对位置拟合到参考位置处,在参考位置等于真实位置时达到最优效果。仿真结果表明,该方法可有效提高定位精度,并具有分布式和可扩展等优点。In 5 G mobile communication system, Device-to-Device(D2 D) communication introduces plenty of internodes’ signal measurements, which can be used to improve application scope and the accuracy of wireless localization. A novel distributed cooperative localization method based on a minimally constrained system(MCS) is proposed. This method adopts the node rough locations derived by existing single node localization methods to construct minimally equality constraints to fix the network location and orientation, and derive the location estimation through by maximizing the likelihood function of the internodes’ measurement. Theoretical investigation indicates that this method is equivalent to fix the relative locations onto some reference locations, which achieves the optimal when the reference location is compose of the ground truth. The simulations conforms the performance of the proposed method and indicates its extendibility and distributed nature.

关 键 词：协作定位 最小约束系统 极大似然估计 相对构型 Procrustes拟合 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]