FFNN-Based Optimization of Embedded Common-Mode Filter for C-PHY
Journal
Final Program - 2026 Asia-Pacific International Symposium and Exhibition on Electromagnetic Compatibility, APEMC 2026
Start Page
1
End Page
3
ISBN
[9798331566593]
Date Issued
2026-05-04
Author(s)
Cheng, Yu-Ying
Abstract
This paper presents a fast design optimization method combining feedforward neural network (FFNN) and particle swarm optimization (PSO) for embedded common-mode filter (CMF) in the MIPI C-PHY interface. The embedded C-PHY CMF is implemented by a defected ground structure (DGS). An alternative FFNN-based model is developed to accurately predict the effect of different CMF structural parameters on the common-mode transmission coefficient (Scc 21). By combining the proposed surrogate model with PSO, the optimal embedded C-PHY CMF structural parameters are efficiently searched to maximize the common-mode noise suppression bandwidth. This method significantly improves the CMF design process with high accuracy and efficiency.
Event(s)
2026 Asia-Pacific International Symposium and Exhibition on Electromagnetic Compatibility, APEMC 2026
Subjects
common-mode filter (CMF)
feedforward neural network (FFNN)
MIPI C-PHY
particle swarm optimization (PSO)
signal integrity (SI)
Publisher
IEEE
Type
conference paper