Rapid Design of Wide-Area Heterogeneous Electromagnetic Metasurfaces Beyond the Unit-Cell Approximation

Abstract

Abstract-We propose a novel numerical approach for the optimal design of wide-area heterogeneous electromagnetic metasurfaces beyond the conventionally used unit-cell approximation. The proposed method exploits the combination of Rigorous Coupled Wave Analysis (RCWA) and global optimization techniques (two evolutionary algorithms namely the Genetic Algorithm (GA) and a modified form of the Artificial Bee Colony (ABC with memetic search phase method) are considered). As a specific example, we consider the design of beam deflectors using all-dielectric nanoantennae for operation in the visible wavelength region; beam deflectors can serve as building blocks for other more complicated devices like metalenses. Compared to previous reports using local optimization approaches, our approach improves device efficiency; transmission efficiency is especially improved for wide deflection angle beam deflectors. The ABC method with memetic search phase is also an improvement over the more commonly used GA as it reaches similar efficiency levels with a 35% reduction in computation time. The method described here is of interest for the rapid design of a wide variety of electromagnetic metasurfaces irrespective of their operational wavelength.

Optimal Design of Beam-Deflectors Using Extended Unit-Cell Metagratings

“…In this paper, we describe a fast numerical optimization technique relying on global optimization methods and semi-analytical forward solvers that can serve as a clear design strategy for designing devices that exploit the synergy between individual and collective effects to realize performance goals. This paper builds on our recent preliminary reports [12,13] and extends the work of Byrnes and coworkers [7] by replacing the local optimization techniques by more appropriate evolutionary optimization techniques. Specifically, we focus our attention to a beam deflector element which is a commonly used structure for bench-marking metasurfaces design strategies [7,14] as the transmission efficiencies can be compared across designs.…”

“…We have added memetic search phase to the ABC algorithm to further improve the search speed. The procedure to determine the correct hyperparameters for the global optimization routines is reported earlier [12]. We have also released the full source code for this implementation, and it is available at [20].…”

Phase Quantized Metasurface Supercells for Wave Manipulation and RCS Reduction

“…The approach of digital elements or codifications of meta-elements according to their phase response which describes behaviors of metamaterial or metasurface is an advantage over effective medium based complex calculations [18][19][20][21]. Implementation of binary codes to represent single/multi-bit elements based on reflection phase discretization can simplify the element distribution over phased reflecting surfaces.…”

“…Analogously, the technique to modify the wave path by phase gradient surfaces led to significant functionalities like the ultrathin flat lens [8,9], orbital angular momentum wavefront generator [10][11][12], polarization converter [13,14], electromagnetic cloaks [15,16], holograms [17], novel antennas [18] and many more. Also, numerical approaches for the efficient and fast design of all-dielectric metasurfaces are proposed recently [19]. Coding metamaterials and digitalization of sub-wavelength element distribution by single/multi-bit optimized sequence attracted many researchers in the recent past.…”

Optimization and uncertainty quantification of gradient index metasurfaces [Invited]

“…However, considering the proximity effects of closely packed antennas is not a simple task and it often requires tedious and time consuming numerical iteration procedures to optimize the overall metasurface response. Several promising works have been recently realized in order to tackle this optimization problem using different optimization methods including for example particle swarm optimization algorithms [30], gradient-based inverse design method [31], adaptive generic [32] and genetic algorithms [33], evolutionary [34] and deep learning models [35]. Although all of these optimization methods could lead to highly efficient designs, their experimental realizations hardly reach the expected performances.…”