论文标题
迭代基于贝叶斯的本地化机制
Iterative Bayesian-based Localization Mechanism for Industry Verticals
论文作者
论文摘要
我们提出并评估使用贝叶斯推断的迭代定位机制，以使用接收的信号强度测量值估算目标的位置。目标坐标的概率密度函数是通过贝叶斯网络估算的。在此，我们考虑了一个迭代过程，每当有新的测量值可用时，我们的预测变量（后验分布）会按顺序更新。根据各自的均方根误差和目标坐标的内核密度估计来评估机制的性能。我们的数值结果表明，提出的迭代机制可以越来越更好地估计目标节点位置，每个更新贝叶斯网络的每个更新都具有新的输入测量。
We propose and evaluate an iterative localization mechanism employing Bayesian inference to estimate the position of a target using received signal strength measurements. The probability density functions of the target's coordinates are estimated through a Bayesian network. Herein, we consider an iterative procedure whereby our predictor (posterior distribution) is updated in a sequential order whenever new measurements are made available. The performance of the mechanism is assessed in terms of the respective root mean square error and kernel density estimation of the target coordinates. Our numerical results showed the proposed iterative mechanism achieves increasingly better estimation of the target node position each updating round of the Bayesian network with new input measurements.
