Physics-Informed Genetic Optimization for Near-Field Beam Shaping in Phased Array Radar Sensing

Area of Science:

• Electromagnetics and Antenna Theory
• Computational Electromagnetics
• Optimization Techniques

Background:

• Near-field beam shaping for phased-array antennas presents a complex electromagnetic synthesis challenge.
• Coherent control of radiated fields is required, considering element positions, patterns, and polarization.
• Existing methods often involve cumbersome coordinate transformations, potentially losing polarization information.

Purpose of the Study:

• To develop a physics-informed optimization framework for near-field beam shaping.
• To enable direct coherent summation of electromagnetic fields from array elements in distinct coordinate systems.
• To optimize electromagnetic energy distribution over finite target surfaces.

Main Methods:

• A unified vector formulation for direct coherent field summation, preserving polarization properties.
• Integration of an analytical near-field point-focusing solution for initialization.
• A genetic algorithm (GA) to solve the non-convex optimization problem for element phase distribution.

Main Results:

• Validated flexible near-field beam shaping and controlled energy focusing over finite target regions.
• Demonstrated the framework's effectiveness through full-wave simulations and experimental measurements.
• Successful design, fabrication, characterization, and calibration of a phased-array platform.

Conclusions:

• The proposed framework offers a rigorous and efficient approach to near-field beam synthesis.
• It overcomes limitations of conventional methods by preserving polarization and simplifying calculations.
• Applicable to diverse fields including biomedical sensing, SAR, wireless power transfer, and millimeter-wave communications.