Artificial intelligence-based 5G network capacity planning and operation

Abstract

The highly demanding requirements envisaged for future 5G networks together with the required support of new customers from vertical industries (e.g. e-health, automotive, energy) pose a big challenge for operators in 5G on how to balance investments, user experience and profitability. There will be the need to revisit the actual methodologies of network planning and operation, fully exploiting cognitive capabilities that embrace knowledge and intelligence to achieve a proper understanding of the network usage in multiple dimensions. In this respect, this paper presents a vision on how these planning and operation processes can rely on the inclusion of Artificial Intelligence (AI) concepts that will allow devising models to characterize the impact of many correlated inputs on specific operator objectives and to drive decisions for different processes.