Research on anti-turbulence methods for vortex beam communication based on FSRA-ConvNet and SFHformer

Area of Science:

• Optical communications
• Signal processing
• Machine learning

Background:

• Vortex light and orbital angular momentum (OAM) enhance spectral efficiency in free-space optical (FSO) communication.
• Atmospheric turbulence causes wavefront distortions, leading to mode dispersion and crosstalk, limiting OAM mode detection and communication reliability.

Purpose of the Study:

• To propose an adaptive demodulator (SFHformer network) for OAM mode demodulation in FSO communication.
• To develop an intensity restoration network (FSRA-ConvNet) to mitigate turbulence-induced wavefront distortions.
• To evaluate the performance of OAM demodulation combined with error correction codes under various turbulence conditions.

Main Methods:

• An SFHformer network was developed as an adaptive demodulator for OAM modes.
• Low-density parity-check (LDPC) encoding and decoding were combined with SFHformer demodulation using MS, NMS, and OMS algorithms.
• An FSRA-ConvNet intensity restoration network, built on UNet with FSRA and Converse2D modules, was proposed to address wavefront distortion.

Main Results:

• The SFHformer-NMS demodulation framework achieved a significant reduction in bit error rate.
• The FSRA-ConvNet effectively restored signal intensity and mitigated wavefront distortions.
• The proposed framework demonstrated high-quality signal transmission capabilities even under extreme atmospheric turbulence.

Conclusions:

• The developed SFHformer-NMS framework offers a robust solution for reliable OAM mode demodulation in FSO systems.
• The FSRA-ConvNet effectively compensates for atmospheric turbulence effects, enhancing signal transmission quality.
• The integrated approach significantly improves the reliability and performance of FSO communication systems in challenging environments.