Augmented wavelength division allocation method for peak traffic grooming in optical networks using 2-layer perceptron learning

Key Points

• Peak traffic grooming was optimized through an augmented wavelength division multiplexing approach, enhancing system efficiency.
• The learning algorithm employed was a 2-layer perceptron, effectively predicting optimal wavelength allocations for traffic patterns.
• Observations indicated significant improvements in data throughput under peak conditions, enhancing overall network performance.
• Implications highlight the potential for improved network design; however, further validation in practical settings is necessary.