.png){kind=icon}
.png){kind=icon}
Return
Efficient RCS Modeling With an Adaptive Design-Based Gaussian Process Method
.png){kind=icon}
PRE
AI
DOI：10.1109/LAWP.2024.3376740.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
An efficient adaptive design-based Gaussian process (GP) method is proposed to predict the radar cross section (RCS) in this letter. To implement effectively supervised learning, an adaptive sampling strategy is presented based on a modified expected improvement for global fit (MEIGF) technique to speed up the convergence with a sequentially augmented training dataset. In particular, the proposed MEIGF strategy can properly capture a compromise position between global exploration and local exploitation. Numerical examples in the prediction of object functions and RCS are calculated to demonstrate the accuracy and efficiency of the proposed method. The results indicate that the proposed method has better performance than the mean squared error and expected improvement for global fit-based traditional GP approaches.
Keywords:
Adaptive sampling
Gaussian process (GP)
modified expected improvement for global fit (MEIGF)
radar cross section (RCS)
Journal
IF:
4.8
Papers:
1.0W
Citations:
2.8W