Motivation
The growing complexity of mobile networks continues to present challenges in optimizing network performance. While 6G promises significant performance improvements, managing 6G operations to ensure optimal performance across diverse environments remains a complex task. In the literature, Machine Learning (ML) techniques have been shown to be effective in optimizing specific aspects of network performance. However, in many cases, multiple aspects of network performance need to be optimized simultaneously. Applying multiple ML models without proper coordination can result in conflicting decisions, potentially causing the network to operate in an undesirable state.
Goal
The project aims to explore solutions for coordinating multiple ML algorithms, develop a unified ML platform for O-RAN, and conduct real-world testing of these algorithms within an O-RAN environment. Initially, the project will focus on developing individual ML algorithms to achieve self-configuration, self-optimization, and self-healing capabilities. Subsequently, these ML models will be integrated, and an ML orchestrator will be designed to manage and resolve conflicting decisions among the various models. Testing will be conducted in two scenarios, targeting applications in both telecommunications and future smart factory environments.
