Quantum Annealing-Inspired Optimization for Space–Time Coding Metasurface

Space-time coding metasurfaces introduce a new degree of freedom in the temporal domain, enabling advanced manipulation of electromagnetic (EM) waves, particularly in controlling waves at different harmonic frequencies. Many applications of such metasurfaces rely on optimization algorithms to achieve specific functionalities. However, the computational cost of these algorithms becomes prohibitive when optimizing metasurfaces with large spatial and time dimensions. To address this challenge, we propose a quantum annealing-inspired optimization framework designed to efficiently optimize space-time coding metasurfaces. First, the scattering behavior of space-time coding metasurface is mapped into the form of a binary spin model, where the phase of each meta-atom, including the discretization into arbitrary bits, is encoded as spins. Next, we construct the fitness function tailored to the desired optimization goals, and the resulting binary spin problem is then solved using a quantum-inspired simulated bifurcation (SB) algorithm. Finally, we demonstrate the effectiveness of our approach through several representative examples, including single-beam steering, multi-beam steering, and waveform design at arbitrary harmonic frequencies. The proposed method significantly enhances the optimization efficiency, delivering high-quality solutions while substantially reducing computational time compared to genetic algorithms (GA), quantum-inspired genetic algorithms (QGA), and simulated annealing (SA). This advancement enables the practical optimization of large-scale space-time coding metasurfaces, paving the way for their broader application in advanced EM wave manipulation.