Energy Efficiency Optimization: A New Trade-off Between Fairness and Total System Performance

TL;DR

This paper introduces a multi-objective energy efficiency optimization method balancing total system performance and fairness, using a low-complexity sequential convex optimization algorithm.

Contribution

It presents a novel multi-objective EE optimization framework that simultaneously considers TEE and MEE, along with a new complexity analysis for SCO algorithms.

Findings

The proposed method achieves various trade-offs between TEE and MEE.

A low-complexity SCO-based algorithm effectively solves the nonconvex optimization problem.

Theoretical results provide insights into the complexity of SCO algorithms.

Abstract

The total energy efficiency (TEE), defined as the ratio between the total data rate and the total power consumption, is considered the most meaningful performance metric in terms of energy efficiency (EE). Nevertheless, it does not depend directly on the EE of each link and its maximization leads to unfairness between the links. On the other hand, the maximization of the minimum EE (MEE), i.e., the minimum of the EEs of all links, guarantees the fairest power allocation, but it does not contain any explicit information about the total system performance. The main trend in current research is to maximize TEE and MEE separately. Unlike previous contributions, this letter presents a general multi-objective approach for EE optimization that takes into account both TEE and MEE at the same time, and thus achieves various trade-off points in the MEE-TEE plane. Due to the nonconvex form of the resulting problem, we propose a low-complexity algorithm leveraging the theory of sequential convex optimization (SCO). Last but not least, we provide a novel theoretical result for the complexity of SCO algorithms.