Optimized Volterra filter equalizer based on weighted principal component analysis for IM-DD transmission.

Area of Science:

• Optical communications
• Signal processing
• Fiber optics

Background:

• Volterra filter equalizers (VFE) are crucial for mitigating impairments in intensity-modulation direct-detection (IM-DD) systems.
• The computational complexity of traditional VFE hinders high-speed, long-reach transceiver applications.

Purpose of the Study:

• To propose a hardware-efficient nonlinear equalizer (NLE) by optimizing the VFE structure.
• To reduce the number of VFE taps without compromising transmission performance.

Main Methods:

• Developed an optimized VFE using weighted principal component analysis (WPCA).
• WPCA combines supervised weights and unsupervised principal component analysis (PCA) for efficient tap reduction.
• Experimental validation in C-band 80 Gb/s and 56 Gb/s PAM-4 transmissions over standard single-mode fiber (SSMF).

Main Results:

• Achieved significant tap reduction in VFE: 40% for 80 Gb/s PAM-4 over 20 km CD-uncompensated SSMF.
• Reduced VFE taps by 60% for 56 Gb/s PAM-4 over 100 km CD-managed SSMF.
• Demonstrated no transmission penalty with the optimized VFE.

Conclusions:

• The proposed WPCA-based VFE offers a hardware-efficient NLE solution for high-speed IM-DD systems.
• Significant tap reduction is achievable, enabling practical implementation in transceivers.
• Validated effectiveness for both CD-uncompensated and CD-managed fiber transmission scenarios.