Base station selection for energy efficient network operation with the majorization-minimization algorithm

TL;DR

This paper proposes a novel majorization-minimization algorithm for energy-efficient base station selection in mobile networks, effectively reducing energy consumption while maintaining service quality, especially in large-scale and heterogeneous networks.

Contribution

It introduces a new suboptimal approach that relaxes the integer programming problem and accounts for inter-cell interference, suitable for large and heterogeneous networks.

Findings

Algorithm achieves low energy consumption while preserving data rates.

Effective in large-scale and heterogeneous network scenarios.

Outperforms existing methods in computational efficiency and applicability.

Abstract

In this paper, we study the problem of reducing the energy consumption in a mobile communication network; we select the smallest set of active base stations that can preserve the quality of service (the minimum data rate) required by the users. In more detail, we start by posing this problem as an integer programming problem, the solution of which shows the optimal assignment (in the sense of minimizing the total energy consumption) between base stations and users. In particular, this solution shows which base stations can then be switched off or put in idle mode to save energy. However, solving this problem optimally is intractable in general, so in this study we develop a suboptimal approach that builds upon recent techniques that have been successfully applied to, among other problems, sparse signal reconstruction, portfolio optimization, statistical estimation, and error correction. More precisely, we relax the original integer programming problem as a minimization problem where the objective function is concave and the constraint set is convex. The resulting relaxed problem is still intractable in general, but we can apply the majorization-minimization algorithm to find good solutions (i.e., solutions attaining low objective value) with a low-complexity algorithm. In contrast to state-of-the-art approaches, the proposed algorithm can take into account inter-cell interference, is suitable for large-scale problems, and can be applied to heterogeneous networks (networks where base station consume different amounts of energy)