Image Classifier Based Generative Method for Planar Antenna Design
This work addresses the challenge of making antenna design accessible to more engineers, particularly in wearable products, by automating the process without requiring prior experience.
The paper tackled the problem of designing planar antennas for PCBs by proposing a method that models antennas with basic components and uses a CNN-based image classifier to determine their positions, resulting in final designs that are realistic and perform comparably to those by experienced engineers.
To extend the antenna design on printed circuit boards (PCBs) for more engineers of interest, we propose a simple method that models PCB antennas with a few basic components. By taking two separate steps to decide their geometric dimensions and positions, antenna prototypes can be facilitated with no experience required. Random sampling statistics relate to the quality of dimensions are used in selecting among dimension candidates. A novel image-based classifier using a convolutional neural network (CNN) is introduced to further determine the positions of these fixed-dimension components. Two examples from wearable products have been chosen to examine the entire workflow. Their final designs are realistic and their performance metrics are not inferior to the ones designed by experienced engineers.