Surrogate-Assisted Framework for SI-Compliant Interconnect Design Optimization Using the Earth Mover's Distance
Researchers have developed a new framework for optimizing the design of SI-compliant PCBs using machine learning and the Earth Mover's Distance (EMD). This approach uses neural surrogate models to predict waveform features and a decision tree to identify compliant waveforms. EMD then ranks these designs based on their similarity to an ideal reference signal, offering a deterministic and interpretable alternative to traditional optimization methods. AI