Broadband Terahertz Diffuse Scattering on Convolutional Coding Metasurfaces

Compared to its counterpart in microwave frequency, which is intensively studied for many years, research of terahertz diffuse scattering is in dire need as the rocketing development of terahertz radar imaging. Coding metasurfaces, labeled by its exotic electromagnetic characteristic in terms of flexibility, programmability, reconfigurability, and computability, are drawing considerable attention increasingly in reductions of the radar cross-section (RCS). Herein, a convolutional encoding strategy is applied on a coding metasurface for broadband enhancement of RCS reduction in the terahertz region. Different from polarization-controlled metasurfaces, the 1-bit coding elements in our meta-device consist of a squared ring resonator and its complementary squared patch resonator with polarization insensitivity and a low Q resonant nature. In avoidance of intercoupling between neighboring meta-atoms, the $4 \times 4$ supercells comprising “0”/“1” binary elements are constructed for the RCS reduction coding array, giving a smooth phase shift of 180° ± 20° and over −0.3 dB reflectances with a fractional bandwidth of 200GHz in 0.28–0.48THz range. Benefiting from convolution calculus of P-type coding and checkerboard coding sequences, the combined merits of both coding methods significantly eliminate the specular reflection and thus enhance the uniform divergence of scattering waves. Consequently, the CST simulation and numerical results exhibit fairly good consistency, corroborating a 25dB RCS reduction with an increment over 5dB compared to each single coding implementation. With a promising perspective, the proposed approach may extend its applications in multifarious scenarios, including radar-signature control, terahertz wireless links, massive MIMO channels, computational imaging, etc.