Beam Shaping of VCSEL Arrays Utilizing Liquid Crystal Spatial Light Modulation

Beam shaping and uniformity are critical requirements in various optical applications, including laser processing, aesthetic treatments, medical therapies, holography, and optical communication. The Gerchberg-Saxton (GS) algorithm and the hybrid input-output (HIO) algorithm are widely used for beam shaping; however, they exhibit notable limitations. While the GS algorithm converges quickly, it is prone to stagnation at local minima. Conversely, the HIO algorithm improves exploration but suffers from instability due to fixed parameter settings. This study proposes an improved phase retrieval algorithm that integrates the strengths of both GS and HIO. It introduces an adaptive damping factor ( $\alpha $ ) that increases over iterations to enhance uniformity while maintaining stability. Additionally, a dynamic interpolation strategy is applied to the $\beta $ parameter to prevent divergence. Numerical simulations confirm the superior performance of the proposed method, achieving a beam uniformity of 88.62% and an energy efficiency of 72.63%, outperforming both the GS and HIO algorithms. This method effectively balances convergence speed, stability, and uniformity, offering a promising solution for high-precision beam shaping. In the experimental validation, the generated grayscale hologram is uploaded to the liquid crystal spatial light modulator (LCSLM), thereby facilitating the homogenization of the vertical-cavity surface-emitting laser (VCSEL) beam. This process yields a rectangular flat-top beam with a measured beam uniformity of 83.85%.