Fast QoT estimation method using cascaded artificial neural network for real-time path provisioning in IMDD based...

Area of Science:

• Optical communications engineering
• Artificial intelligence in telecommunications
• Signal processing for optical networks

Background:

• Quality of Transmission (QoT) estimation is crucial for optimizing optical network performance.
• Traditional methods for QoT estimation can be computationally intensive and slow.
• Intensity Modulation-Direct Detection (IMDD) systems are widely used in short-reach optical communication.

Purpose of the Study:

• To propose a computationally efficient method for QoT estimation in IMDD systems.
• To accurately predict the bit error rate (BER) considering deterministic waveform distortions.
• To enable real-time performance monitoring and network management.

Main Methods:

• Development of a cascaded Artificial Neural Network (ANN) model.
• Incorporation of deterministic waveform distortion models, including chromatic dispersion and self-phase modulation (SPM).
• Validation of the ANN model for a three-span 36 km transmission link.

Main Results:

• The proposed ANN-based method achieves fast QoT estimation, completing calculations within 0.7 seconds.
• The method accurately accounts for signal impairments like chromatic dispersion and SPM.
• Achieved high accuracy in predicting bit error rates for the simulated transmission scenario.

Conclusions:

• Cascaded ANNs offer a viable and efficient solution for real-time QoT estimation in IMDD systems.
• The proposed method enhances the capability for dynamic network management and resource allocation.
• This approach contributes to the development of more robust and performant optical communication networks.