Low-cost yield-driven design of antenna structures using response-variability essential directions and parameter...

Area of Science:

• Electrical Engineering
• Electromagnetics
• Antenna Theory

Background:

• Fabrication tolerances and material uncertainties significantly impact antenna performance, causing issues like frequency shifts and impedance mismatch.
• Maximizing fabrication yield is crucial for antenna design, especially in industrial applications, but traditional methods are computationally expensive.
• Existing surrogate modeling techniques struggle with the curse of dimensionality, limiting their effectiveness in complex antenna design optimization.

Purpose of the Study:

• To propose a novel, low-cost technique for yield optimization of antenna structures considering fabrication tolerances.
• To reduce the computational burden associated with traditional electromagnetic (EM) analysis and surrogate modeling in antenna design.
• To enhance antenna design immunity to manufacturing inaccuracies and material parameter variations.

Main Methods:

• Developed a technique that defines a limited domain for metamodel construction, focusing on essential directions influencing antenna responses.
• Utilized an automated decision-making process to identify these influential directions based on circuit response variability.
• Validated the proposed method using three antenna structures and compared it against benchmark surrogate-assisted methods.

Main Results:

• The proposed approach maintains a small domain volume, leading to lower surrogate model setup costs.
• Achieved significant yield improvement for antenna structures despite reduced model complexity.
• Demonstrated the reliability of the yield optimization process through electromagnetic-driven Monte Carlo simulations.

Conclusions:

• The presented technique offers an efficient and cost-effective solution for antenna yield optimization under manufacturing uncertainties.
• This method effectively mitigates the curse of dimensionality problem in surrogate modeling for antenna design.
• The findings are crucial for improving the robustness and manufacturability of antennas in industrial contexts.