AI-Driven Optimization in Free-Space Optical Communications

Free-space optical (FSO) communication has emerged as a promising technology for next-generation wireless and optical communication networks due to its ultra-high bandwidth, low latency, enhanced security, and license-free spectrum operation. FSO systems are increasingly being investigated for applications such as 6G networks, satellite and inter-satellite links, unmanned aerial vehicle (UAV) communication, and integrated space–air–ground architectures. However, the practical deployment and consistent operation of FSO systems remain highly vulnerable to atmospheric turbulence, fog, rain, haze, beam wandering, pointing errors, and link misalignment, all of which can significantly degrade reliability and stability under dynamic conditions. Recent progress in artificial intelligence (AI), machine learning (ML), and deep learning (DL) has opened new possibilities for intelligent and adaptive FSO communication systems capable of responding to environmental and mobility variations in real time. Existing studies demonstrate that AI-driven approaches can enhance system performance through adaptive modulation and coding, intelligent beam alignment and tracking, predictive outage management, autonomous link control, and learning-based or physics-informed channel estimation. Furthermore, hybrid FSO/radio-frequency (RF) architectures with intelligent switching and allocation mechanisms have shown strong potential to maintain connectivity during adverse weather and temporary link degradation. Despite this progress, reproducible and experimentally validated AI solutions are still needed to ensure robust performance in realistic, dynamic conditions.



This Research Topic aims to explore how artificial intelligence can enhance the robustness, adaptability, and operational reliability of free-space optical communication systems under diverse atmospheric and mobility conditions. The goal is to advance methods that combine learning-based adaptability with physical modeling to mitigate environmental impairments such as atmospheric turbulence, beam misalignment, and weather-driven loss of signal integrity. The Topic encourages research into AI-assisted modulation and coding, predictive link management, and autonomous optimization frameworks that enable FSO systems to sustain reliable performance across variable node mobility and environmental states. Equally important are lightweight, hardware-efficient, and energy-conscious AI architectures suited for edge deployment and low-latency operation. The Topic places particular emphasis on interpretable, trustworthy, and experimentally substantiated AI techniques that can ensure reproducibility, transparency, and technical feasibility for next-generation intelligent optical wireless networks.



To gather further insights into AI-powered resilience for FSO communication, we welcome contributions presenting original research, datasets, simulation studies, hardware experiments, and perspectives that propose or validate innovative and practical solutions for intelligent optical wireless communication. Submissions emphasizing deployment-aware design, hardware-in-the-loop experimentation, reproducible benchmarking, and evaluation across realistic atmospheric conditions are particularly encouraged. We welcome contributions addressing, but not limited to, the following themes:



- AI-assisted atmospheric turbulence mitigation

- Adaptive modulation, coding, and resource allocation

- Intelligent beam steering, tracking, and alignment

- Predictive and autonomous link management

- Hybrid FSO/RF communication and switching strategies

- Reinforcement learning for autonomous communication optimization

- Learning-based and physics-informed channel estimation

- Explainable and trustworthy AI for optical wireless systems

- Weather-aware communication frameworks

- Edge AI and lightweight AI architectures for real-time control

- Digital twin technologies for intelligent optical wireless systems

- Hardware-in-the-loop testing and experimental validation

- Reproducible evaluation methodologies under realistic atmospheric conditions



Contributions involving experimental demonstrations, outdoor optical wireless experiments, digital twins, hardware-aware implementations, and practical deployment investigations under realistic environmental conditions are particularly encouraged.