Random-phase metasurface-assisted optical phased arrays for large-FoV and high-resolution beam steering

Optical phased arrays (OPAs) are promising beam-steering devices for a variety of applications, including light detection and ranging (LiDAR), wireless optical communication (OWC), optical imaging, and biomedical sensing. Current OPAs suffer from a limited field of view (FoV) due to grating lobes. In addition, the OPAs technology faces challenges such as the complex design of antenna arrays and the need for high-precision and widely tunable integrated lasers for one-dimensional (1D) OPAs. In this study, we propose a method using a passive random-phase metasurface to reconstruct the optical field of OPAs, enabling grating-lobe suppression, broadening of the element radiation pattern, and improved resolution, which together result in high-resolution beam steering over a wide field of view. A theoretical model is established to explain the working principle of the proposed method and to analyze its beam-steering performance. The model reveals quantitative relationships between performance metrics, including the side lobe suppression ratio (SLSR), resolution, and efficiency, and system parameters such as the number of OPA channels and the propagation distance. In addition, numerical simulations are conducted to validate the theoretical model and analyze the validity of the model. Moreover, a 2D beam steering is demonstrated numerically using a 1D OPA combined with a random-phase metasurface, without lenses or wavelength tuning. Continuous beam steering is enabled by the optical memory effect. Finally, the impact of practical meta-atom modeling on system performance is investigated. This work introduces a practical, low-cost, and scalable approach to enhancing OPA beam-steering performance using a random-phase metasurface. The presented study provides design guidelines for implementing this method in advanced integrated OPA-metasurface systems.