Technology Category
Search and Optimization: Non-convex OptimizationCognitive Modeling & Cognitive Systems: Neural Spike CodingMachine Learning: Hardware-aware ML
Nonlinear Distortion Equalization in Multi-Span Optical Links Via a Feed-Forward Photonic Neural Network
🤖 AI Summary
To address signal degradation caused by linear (e.g., chromatic dispersion) and nonlinear (e.g., self-phase modulation—SPM—and cross-phase modulation—XPM) distortions in multi-span optical links, this paper proposes and experimentally demonstrates a silicon-photonic integrated feedforward photonic neural network (PNN) equalizer. The PNN combines an adjustable 8-tap optical FIR filter with square-magnitude nonlinear operations to achieve low-latency, low-power all-optical compensation. Employing an intensity-modulation/direct-detection (IM/DD) architecture, the system integrates a tunable dispersion compensator and edge-coupled photodetectors. Experimentally, it achieves efficient equalization of 10-Gbps signals under cumulative dispersion of 200 km and SPM-induced distortion equivalent to 450 km of standard single-mode fiber. Numerical simulations further indicate scalability to 100-Gbps operation and capability to mitigate XPM impairments. This work establishes a compact, energy-efficient on-chip nonlinear equalization paradigm for high-speed short-reach optical interconnects.
Technology Category
Application Category
📝 Abstract
Linear and nonlinear distortions in optical communication signals are equalized using an integrated feed-forward Photonic Neural Network (PNN). The PNN is based on a linear stage made of an 8-tap Finite Impulse Response (FIR) filter, featuring tunable amplitude and phase weights at each tap, and of a nonlinear stage achieved through the square modulus operation at the end-of-line photodetector. Within an Intensity Modulation/Direct Detection (IMDD) system, the PNN is applied to 2-level Pulse Amplitude Modulated (PAM2) optical signals undergoing multi-span propagation. Each 50 km segment includes fiber transmission, optical power restoration, and optional chromatic dispersion compensation via a Tunable Dispersion Compensator. Positioned at the receiver, the PNN enables fully optical signal processing with minimal latency and power consumption. Experimental validation is conducted using a Silicon-On-Insulator device operating on 10 Gbps signals. It demonstrates chromatic dispersion equalization over distances up to 200 km and self-phase modulation (with dispersion removed) up to 450 km. Simulations explore PNN adaptation for 100 Gbps modulations and its potential for cross-phase modulation equalization.
Problem
Research questions and friction points this paper is trying to address.
Equalizes linear and nonlinear distortions in optical signals
Enables low-latency optical processing for multi-span links
Validates dispersion and modulation compensation up to 450 km
Innovation
Methods, ideas, or system contributions that make the work stand out.
Integrated feed-forward Photonic Neural Network (PNN)
8-tap FIR filter with tunable weights
Square modulus operation for nonlinear stage
🔎 Similar Papers
2024-09-05Journal of Lightwave TechnologyCitations: 0
