Free-space optical (FSO) communication provides high-capacity and secure links but is strongly impaired by atmospheric turbulence, which induces multi-scale irradiance fluctuations. Traditional approaches such as adaptive optics, multi-aperture and multiple-input multiple-output FSO schemes offer limited robustness under rapidly varying turbulence, while statistical fading models such as log-normal and Gamma–Gamma cannot represent multi-scale temporal correlations. This work proposes a hybrid deep learning framework that explicitly separates high-frequency scintillation and low-frequency power drift through a conditional variational autoencoder and a bidirectional long short-term memory dual-branch architecture with an adaptive gating mechanism. Trained on OptiSystem-generated datasets, the model accurately reconstructs irradiance distributions and temporal dynamics. For model-assisted signal compensation, it achieves an average 79% bit-error-rate (BER) reduction across all simulated scenarios compared with conventional thresholding and Gamma–Gamma maximum a posteriori detection. Transfer learning further enables efficient adaptation to new turbulence conditions with minimal retraining. Experimental validation shows that the compensated BER approaches near-zero, yielding significant improvement over traditional detection. These results demonstrate an effective and adaptive solution for turbulence-impaired FSO links.

This paper proposes a physics-guided hybrid control framework for real-time suppression of micro-vibration-induced beam jitter in long-distance LEO-GEO (low Earth orbit-geostationary Earth orbit) optical communication links. Building on a disturbance-observer-based Linear Quadratic Integral architecture with output-only autoregressive identification, the approach removes the need for explicit disturbance input measurements while preserving modeling accuracy. A lightweight neural module, implemented via a custom-designed Transformer model, is incorporated as a bounded residual compensator to refine control performance under non-stationary disturbances, and is further safeguarded by a gating mechanism to ensure stable fallback operation. Experiments on a production-grade optical terminal under representative multi-condition disturbances show consistently high identification correlation (up to 0.9997) and a substantial reduction in RMS jitter from 0.324918 to 0.04841 µ rad, with only 0.3% additional computational overhead, outperforming conventional control approaches. These results support reliable and efficient deployment for high-precision beam stabilization in spaceborne optical systems.