Generalized Optical Signal Processing Based on Multioperator Metasurfaces Synthesized by Susceptibility Tensors

This paper theoretically proposes a multichannel nonlocal metasurface computer characterized by generalized sheet transition conditions (GSTCs) and surface susceptibility tensors. The study explores polarization- and angle-multiplexed metasurfaces enabling multiple and independent parallel analog spatial computations when illuminated by differently polarized incident beams from different directions. The proposed synthesis overcomes substantial restrictions imposed by the previous designs such as large architectures arising from the need for additional sub-blocks; slow responses; working for a certain incident angle or polarization; executing only a single mathematical operation; and, most importantly, supporting only the even-symmetric operations for normal incidences. The versatility of our design is demonstrated in a way that an ultracompact, integrable, and homogeneous metasurface-assisted platform can execute a variety of optical-signal-processing operations such as spatial differentiation and integration. It is demonstrated that a metasurface featuring nonreciprocal properties can be thought of as a new paradigm to break the even symmetry of reflection and perform both even- and odd-symmetric mathematical operations for input fields coming from a normal direction. Numerical simulations also illustrate different aspects of a multichannel edge detection scheme through projecting multiple images on the metasurface from different directions. Such appealing findings not only circumvent the major potential drawbacks of previous designs but may also offer an efficient, easy-to-fabricate, and flexible approach in wave-based signal processing, edge detection, image contrast enhancement, hidden object detection, and equation solving without a Fourier lens.

Parallel integro-differential equation solving via multi-channel reciprocal bianisotropic metasurface augmented by normal susceptibilities

Karim Achouri, Ali Momeni

Analog optical signal processing has dramatically transcended the speed and energy limitations accompanied with its digital microelectronic counterparts. Motivated by recent metasurface?s evolution, the angular scattering diversity of a reciprocal passive bianisotropic metasurface with normal polarization is utilized in this paper to design a multi-channel meta-computing surface, performing multiple advanced mathematical operations on input fields coming from different directions, simultaneously. Here, the employed ultra-thin bianisotropic metasurface computer is theoretically characterized based on generalized sheet transition conditions and susceptibility tensors. The operators of choice are deliberately dedicated to asymmetric integro-differential equations and image processing functions, like edge detection and blurring. To clarify the concept, we present several illustrative simulations whereby diverse wave-based mathematical functionalities have been simultaneously implemented without any additional Fourier lenses. The performance of the designed metasurface overcomes the nettlesome restrictions imposed by the previous analog computing proposals such as bulky profiles, asserting only single mathematical operation, and most importantly, supporting only the even-symmetric operations for normal incidences. Besides, the realization possibility of the proposed metasurface computer is conceptually investigated via picturing the angular scattering behavior of several candidate meta-atoms. This work opens a new route for designing ultra-thin devices executing parallel and accelerated optical signal/image processing.

IOP PUBLISHING LTD2019

Parallel Analog Computing Based on a 2×2 Multiple-Input Multiple-Output Metasurface Processor With Asymmetric Response

Amirhossein Babaee, Romain Christophe Rémy Fleury, Ali Momeni

We present a polarization-insensitive metasurface processor to perform spatial asymmetric filtering of an incident beam, thereby allowing for real-time parallel analog processing. To enable massive parallel processing, we introduce a multiple-input multiple-output (MIMO) computational metasurface with asymmetric response that can perform spatial differentiation on two distinct input signals regardless of their polarization. In our scenario, two distinct signals set in x and y directions, parallel and perpendicular to the incident plane, illuminate simultaneously the metasurface processor, and the resulting differentiated signals are separated from each other via appropriate spatial low-pass filters. By leveraging generalized sheet transition conditions and surface susceptibility tensors, we design an asymmetric meta-atom augmented with normal susceptibilities to reach asymmetric response at normal beam illumination. Proof-of-principle simulations are also reported along with the successful realization of signal processing functions. The proposed metasurface overcomes major shortcomings imposed by previous studies, such as large architectures arising from the need for additional subblocks, slow responses, and, most importantly, supporting only a single input with a given polarization. Our results set the path for future developments of material-based analog computing using efficient and easy-to-fabricate MIMO processors for compact, fast, and integrable computing elements without any Fourier lens.

2021