AI-Driven Optimization for FCBGA PKG RF SI

This paper presents an AI-driven optimization approach using Cadence Optimality to enhance the RF performance of FCBGA packages. By refining package routing and via designs, we achieved ~4dB margin improvement in return loss and a reduction of insertion loss by ~0.8dB at high frequency. Additionally, comprehensive link cascading and process variation validations were conducted to ensure the robustness of the optimized design. The results demonstrate AI tool can help significant advancements in signal integrity, providing valuable insights for future high-frequency electronic device packaging.