ENERGY EFFICIENCY OPTIMIZATION IN MASSIVE MIMO USING A PRICING BASED APPROACH

Abstract

Massive Multiple-Input Multiple-Output (mMIMO) is a promising technology for improving the energy efficiency (EE) of wireless communication networks. However, there are several conflicting objective problems that need to be addressed to achieve optimal EE in mMIMO network. In this paper, we present an optimization framework, which is based on multi-objective optimization (MOO) used to investigate a pricing cost based approach in mMIMO network. A network pricing cost is introduced to the energy consumption as a penalty for the achievable spectral efficiency (SE), and studies its impact on the tradeoff between the EE and the throughput. It is difficult to directly solve the problem as it is non-convex, and thus scalarization technique was used to transform the MOO problem into a single-objective optimization (SOO) instead of using iterative algorithms that relied on the dual decomposition for obtaining the pricing cost, optimal throughput and EE. Finally, numerical simulations are used to characterize the interaction between the EE and throughput for various network parameters such as power when the network is designed from the energy-efficient perspective.