Artificial Intelligence in 5G Planning: Optimization of EnodeB Planning Based on 4G KPIs

Deploying a 5G network without considering the existing network infrastructure can be a lengthy and expensive process. This approach requires a thorough evaluation on economic, strategic, and technical fronts, potentially resulting in a significant increase in deployment budget. That’s why, during the initial phase of 5G deployment, it’s common to leverage the existing 4G core infrastructure. In this study, we introduce an approach aimed at optimizing the planning of enodeBs for a 5G mobile network by utilizing statistical data from the existing 4G network usage.Our method relies on the use of artificial intelligence (AI) to automate the selection of a subset of areas to be included in the deployment of the 5G network from a larger set of available zones, with the aim of maximizing customer coverage. To achieve this, we have integrated a classical genetic algorithm. This approach was chosen due to its significant advantages, including its ability to converge towards an optimal solution, even when starting from a solution that is very far from the optimum. Subsequently, practical tests were conducted to assess the effectiveness of our algorithm, and the results clearly demonstrated its efficiency.