Improved SPGD Algorithm for Optical Phased Array Phase Calibration

A chip-level optical beam steerer is an inevitable choice for next-generation light detection and ranging (LiDAR). The research on phased array (OPA) the most intriguing. However, complexity of control calibration speed limit full potential as number channels increases. In this paper, improved stochastic parallel gradient-descent algorithm combined with Nesterov accelerated gradient method (NSPGD) presented applied in a 512-channel OPA. This can reduce phase time large-scale OPA demonstrates better convergence performance than traditional SPGD. Compared SPGD hill-climbing (HC) algorithm, optimized NSPGD shown. side mode suppression ratio (SMSR) over 10dB obtained twice that addition, temperature-controlled also studied to stabilize whole system.