End-to-end constellation shaping for fiber-terahertz integrated communications

Area of Science:

• Optical Communications
• Wireless Communications
• Signal Processing

Background:

• 6G wireless systems face challenges with linear and nonlinear distortions in fiber-terahertz links, limiting spectral efficiency.
• High spectral efficiency is crucial for advanced 6G applications like wireless fronthaul and access networks.

Purpose of the Study:

• To propose and demonstrate an end-to-end four-dimensional joint geometric and probabilistic shaping (E2E-4D-JGPS) framework for fiber-terahertz communication systems.
• To improve data rates and tolerance to distortions in 6G integrated communication systems.

Main Methods:

• Developed a constellation-level equalized channel modeling method using a mixture density network (MDN).
• Implemented generalized mutual information (GMI) as an optimization metric to adaptively design constellation diagrams.
• Utilized an end-to-end framework connecting constellation shaping and de-mapping.

Main Results:

• The proposed 4D-OS1024-PS constellation achieved over 1.68 dB sensitivity gain at a GMI threshold of 8.5 bit/4D-sym.
• Demonstrated a data rate increase to 16 Gbps in a 300 GHz independent-sideband (ISB) modulated 2x2 MIMO system.
• The E2E-4D-JGPS framework showed improved tolerance to nonlinear distortions and noise compared to traditional 2D-32QAM.

Conclusions:

• E2E-4D-JGPS effectively enhances spectral efficiency and performance in fiber-terahertz communication systems.
• The developed method offers a promising solution for overcoming distortion challenges in 6G wireless networks.
• This approach contributes to the advancement of high-capacity, high-performance integrated communication systems.