Related Experiment Video
Updated: Aug 16, 2025

Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
Published on: March 20, 2017
Probabilistic neural network equalizer for nonlinear mitigation in OAM mode division multiplexed optical fiber
A new probabilistic neural network (PNN) equalizer effectively reduces nonlinear impairments in orbital angular momentum (OAM) mode-division multiplexing (MDM) fiber communications. This PNN equalizer enhances receiver sensitivity and lowers computational complexity for ultra-high-capacity systems.
Area of Science:
- Optical Communications
- Signal Processing
- Machine Learning
Background:
- Orbital angular momentum (OAM) mode-division multiplexing (MDM) is crucial for achieving ultra-high-capacity optical fiber communications.
- Nonlinear impairments from optoelectronic devices pose a significant challenge for OAM-MDM systems.
- Existing equalization techniques struggle to fully mitigate these stochastic nonlinear distortions.
Purpose of the Study:
- To present and evaluate a probabilistic neural network (PNN) equalizer for mitigating nonlinear impairments in OAM-MDM systems.
- To demonstrate the effectiveness of the PNN equalizer in improving receiver sensitivity and reducing computational complexity.
- To explore the potential of PNN for high-speed intensity-modulation direct-detection (IM-DD) OAM-MDM fiber communication systems.
Main Methods:
- An equalizer based on a probabilistic neural network (PNN) was designed and implemented.
- The PNN equalizer utilizes Bayesian decision theory to calculate nonlinearity distributions and mitigate stochastic impairments.
- The system employed 32 GBaud Nyquist pulse amplitude modulation-8 (PAM8) intensity-modulation direct-detection (IM-DD) signals in an OAM-MDM fiber communication setup.
Main Results:
- The PNN equalizer significantly mitigated nonlinear signal distortions in the OAM-MDM system.
- Compared to a convolutional neural network (CNN) equalizer, the PNN equalizer improved receiver sensitivity by 0.6 dB (OAM mode l=+3) and 2 dB (OAM mode l=+4) at the 20% forward error correction (FEC) limit.
- The PNN equalizer demonstrated substantially reduced computational complexity compared to Volterra or CNN equalizers.
Conclusions:
- The PNN equalizer is a highly effective method for mitigating nonlinear impairments in high-speed OAM-MDM fiber communication systems.
- The PNN approach offers superior receiver sensitivity and lower computational cost compared to existing equalization techniques.
- This technology holds significant potential for advancing the capacity and efficiency of future optical communication networks.
Related Concept Videos
Propagation of Uncertainty from Random Error
Reducing Line Loss
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
Propagation of Uncertainty from Systematic Error
Linear Approximation in Frequency Domain
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
Network Function of a Circuit
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...

