Abstract
This paper presents the results of a joint investigation conducted between academia and industry. The investigation focused on modeling and reducing the energy expenditure of a 4G LTE network in a London case-study area, using data sets from an existing 3G network deployment. In the first part of the paper, different multi-cell network modeling approaches were compared, including: stochastic and linear geometry, hexagonal cell layout with wrap-around, and realistic network topology. The impact of terrain and clutter is also examined. It was found that they had a similar performance profile (80%) correlation, and a back-off factor to translate theoretical to measured results can account for most of the differences. The second part of the paper focuses on effective techniques that can be used to reduce energy consumption. Sleep mode and heterogeneous networks are combined and an ERG of 15-46% is achieved.