Energy and cost implications of a traffic aware and quality-of-service constrained sleep mode mechanism

This study considers the theoretical and realistic performance of a traffic aware sleep mode mechanism in the cellular network. The study devises a sleep mode mechanism that maximises the energy saving of the network, while maintaining both the user throughput and reliability performance. The analysis combines recent developments in stochastic geometry and packet data modelling, as well as a simulation of a real cellular network. The main contribution is to show that accurate theoretical modelling of realistic network data can allow deterministic sleep mode triggers to be devised. This is shown to be highly applicable to a real network, reducing energy consumption by 15-60% throughout a day and the annual operational expenditure by 4%.