RQ-VAE-based semantic comunication for image transmission in an N-D CAP underwater wireless optical communication...

Area of Science:

• Optical Communications
• Signal Processing
• Data Transmission

Background:

• Underwater wireless optical communication (UWOC) offers high-speed data transmission but struggles with efficiency in challenging environments.
• Traditional UWOC focuses on bit error rate reduction, which is often inefficient.
• Semantic communication prioritizes faithful reconstruction of information over exact bit recovery, improving robustness.

Purpose of the Study:

• To propose a novel semantic communication framework for underwater image transmission.
• To enhance the efficiency and robustness of UWOC systems.
• To achieve higher data transmission rates for underwater imagery.

Main Methods:

• Developed a semantic communication framework using a residual quantization variational autoencoder (RQ-VAE).
• Integrated N-dimensional carrier-less amplitude phase (N-D CAP) modulation for multi-user concurrent transmission.
• Implemented the framework in a 450 nm laser-based UWOC system over a 5 m distance, comparing against JPEG and LDPC standards.

Main Results:

• The proposed semantic communication scheme significantly outperformed traditional JPEG and LDPC standards.
• Achieved up to 4.68 dB improvement in Peak Signal-to-Noise Ratio (PSNR) at similar frame rates.
• Demonstrated a 329% increase in frame rate at similar PSNR.
• Reached a transmission rate of 4,394,531 image instances/patches per second (64x64 resolution).

Conclusions:

• The semantic communication framework based on RQ-VAE and N-D CAP offers superior performance for underwater image transmission.
• This approach substantially reduces data volume and enhances transmission efficiency and robustness.
• The achieved transmission rate represents a significant advancement in UWOC technology for underwater imagery.