Developments of Machine Learning Solution for Assignment of Calls of RWA in Wireless Network

Abstract

The increasing demand for efficient resource management in wireless networks has necessitated innovative solutions for the Routing and Wavelength Assignment (RWA) problem. Traditional approaches often face limitations in handling dynamic traffic conditions, leading to suboptimal resource utilization and increased blocking probabilities. This study explores the development of machine learning-based solutions for the assignment of calls in RWA, offering a data-driven approach to optimize resource allocation and improve network performance. Machine learning algorithms, including supervised learning, reinforcement learning, and neural networks, are employed to predict traffic patterns, dynamically allocate wavelengths, and select optimal routing paths. These models adapt to changing network conditions, enabling real-time decision-making and reducing call blocking. Simulation results demonstrate that machine learning techniques outperform conventional heuristic methods in terms of blocking probability, resource utilization, and adaptability to network fluctuations. This research highlights the potential of machine learning to transform RWA in wireless networks, providing scalable and intelligent solutions for modern communication systems. The findings lay the groundwork for further exploration into hybrid models, cross-domain applications, and integration with emerging technologies such as 5G and beyond.