A bi-objective mixed integer non-linear programming model for the cell switch-off problem considering quality of service and energy consumption

Abstract

In an increasingly interconnected world, there is an emerging demand for new applications and services from wireless networks that are therefore under continuous technological developments to cope with them. One of the enabling technologies for the next generations of these networks (5G/6G) lies in deploying a high number of base stations in the working area, which leads to what is known as ultra-dense networks. This densification allows for enhanced network performance (data rates, latency, etc.), but it also induces a notable increase in the energy consumption of the entire system. One of the pivotal strategies to reduce such a consumption is to switch off cells in periods of low traffic demand. This decision problem is formulated as an optimization problem known as Cell Switch Off (CSO). In this context, this work introduces a bi-objective mixed integer non-linear programming model to address the CSO problem. The primary objectives of this model are twofold: to ensure a high-quality service for end-users and to minimize energy consumption within the network infrastructure. The model was tested in a preliminary computational experimentation over realistic instances allowing to explore the trade-off among the different objectives.