Efficient algorithm for resource optimization in optical communication networks.

Area of Science:

• Telecommunications Engineering
• Wireless Communication Systems
• Machine Learning Applications

Background:

• Next-generation communication systems (beyond 5G/6G) require high throughput, low latency, high dependability, and energy efficiency.
• Hybrid optical-RF communication systems offer a promising approach to meet these demanding requirements.
• Federated Learning (FL) enables distributed machine learning (ML) on smart devices without compromising data privacy.

Purpose of the Study:

• To propose a novel resource optimization solution for hybrid optical-RF communication networks.
• To enhance network efficiency by optimizing user selection, transmission power, and channel estimation.
• To leverage multilayer perceptron and joint optimization techniques for improved network performance.

Main Methods:

• Development of a resource optimization algorithm based on multilayer perceptron.
• Optimization of user selection, transmission power, and channel estimation parameters.
• Joint optimization of user selection and transmission power to minimize the loss function.

Main Results:

• The proposed algorithm demonstrates superior performance compared to existing methods in optical-RF networks.
• Effective optimization of network resources leading to improved system efficiency.
• Successful application of machine learning for resource management in hybrid communication systems.

Conclusions:

• The integration of optical and RF systems, coupled with advanced ML techniques like FL, is crucial for future communication networks.
• The proposed multilayer perceptron-based optimization significantly enhances resource allocation in hybrid optical-RF networks.
• This approach offers a viable solution for achieving high throughput, low latency, and energy efficiency in next-generation wireless systems.