Optimal array design of deficient coding metasurfaces for terahertz beam steering via phase-gradient approaching

Area of Science:

• Metasurfaces and Nanophotonics
• Electromagnetics and Wave Propagation
• Terahertz Technology

Background:

• Terahertz active coding metasurfaces face limitations in phase shift range and precision, hindering beam-steering.
• Existing optimization methods are often computationally intensive and less effective for deficient metasurfaces.

Purpose of the Study:

• To develop an enhanced array-level optimization framework for terahertz beam steering.
• To improve the performance of 2-bit coding metasurfaces using a novel optimization approach.

Main Methods:

• An array-level optimization framework combining phase-coding modulation and genetic algorithm (GA).
• A phase-gradient approaching (PGA) algorithm to generate a near-optimal initial population for GA.
• Design and simulation of a passive, 2-bit coding metasurface at 220 GHz.

Main Results:

• The PGA method with 150° phase modulation significantly increased main lobe amplitude (avg. 2.29 dB) and reduced sidelobe level (avg. 2.52 dB) compared to fractional coding.
• PGA-initiated GA optimization reduced computational iterations by 61.08% compared to fractional coding optimization.
• Demonstrated enhanced beam steering under normal and oblique incidences.

Conclusions:

• The proposed PGA-initiated GA optimization framework offers a feasible solution for enhancing non-ideal coding metasurfaces.
• This method significantly improves beam-steering performance and computational efficiency.
• Potential applications include terahertz wireless communication, computational imaging, and compressive sensing.