公开(公告)号：KR101333111B1

公开(公告)日：2013-11-26

申请号：KR1020120129266

申请日：2012-11-15

Applicant: 국방과학연구소

Inventor： 김갑진 ,  임혁 ,  오벨레,브라운슨.오. ,  김량수 ,  황선녕

IPC: G01S5/14

Abstract: The present invention relates to technology for improving accuracy in wireless positioning and, more specifically, to a system and a method capable of implementing a probabilistic inference based on a Bayesian network model for improving the accuracy in the wireless positioning which uses received signal strength (RSS), time of arrival (ToA), and time difference of arrival (TDoA) information fusion. The present invention is operated by the fusion of a multiple wireless positioning method with the probabilistic inference based on the Bayesian network model, thereby facilitating stable positioning by using a set of information obtained from other positioning methods in case that the obtainment of accurate information is difficult in one positioning system by a malicious external attack. [Reference numerals] (AA) Start;(BB) End;(S600) Assume a prior probability distribution regarding initial random variables;(S610) Input RSS, ToA, TDoA information collected from each reference node;(S620) Extract a Gibbs sample by applying Gibbs Sampler algorithm;(S630) Estimate the location of a target node based on the extracted Gibbs sample

Abstract translation: 本发明涉及用于提高无线定位精度的技术，更具体地，涉及一种能够实现基于贝叶斯网络模型的概率推理的系统和方法，用于提高使用接收信号强度的无线定位精度（RSS ），到达时间（ToA）和到达时间差（TDoA）信息融合。 本发明通过多重无线定位方法与基于贝叶斯网络模型的概率推理的融合来操作，从而通过在获得准确信息困难的情况下使用从其他定位方法获得的一组信息来促进稳定定位 在一个定位系统中通过恶意的外部攻击。 （S6）从每个参考节点收集的输入RSS，ToA，TDoA信息;（S620）提取吉布斯样本（参考数字）（AA）开始;（BB）结束;（S600） 通过应用Gibbs Sampler算法;（S630）基于提取的吉布斯样本估计目标节点的位置