Low-complexity clustered raised-cosine-optimized FIR time-domain digital back-propagation for coherent optical...

Area of Science:

• Optical Communications
• Digital Signal Processing

Background:

• Digital back-propagation (DBP) is crucial for compensating impairments in coherent optical systems.
• Existing methods like FIR TD-DBP and FD-DBP face limitations in accuracy and computational complexity.
• Chromatic dispersion (CD) and nonlinear impairments significantly degrade signal quality.

Purpose of the Study:

• To develop a low-complexity time-domain digital back-propagation (TD-DBP) algorithm for joint CD and nonlinear impairment compensation.
• To overcome accuracy degradation in FIR TD-DBP and high computational overhead in FD-DBP.
• To improve the efficiency of digital nonlinear equalization in coherent optical systems.

Main Methods:

• Proposed a clustered RC-FIR TD-DBP algorithm integrating RC-optimized FIR filtering and k-means++ coefficient clustering.
• Employed an optimized RC window function to enhance tap coefficients and suppress out-of-band spectral leakage.
• Utilized k-means++ clustering to reduce redundant filter coefficients and computational complexity.

Main Results:

• Simulations showed a 55.5% complexity reduction for clustered RC-FIR TD-DBP compared to non-clustered FIR TD-DBP, with a minimal Q-factor penalty.
• The proposed algorithm demonstrated 33.3% lower complexity than FD-DBP with comparable transmission quality in a 160 Gb/s system.
• Experimental results confirmed significant complexity reduction (58.0%) with negligible Q-factor penalty in an 80 Gb/s system.

Conclusions:

• The clustered RC-FIR TD-DBP algorithm effectively compensates for CD and nonlinear impairments with reduced computational complexity.
• The proposed method overcomes the trade-off between computational efficiency and compensation accuracy in digital nonlinear equalization.
• This approach offers a practical solution for enhancing performance in high-speed coherent optical communication systems.