Performance Modeling and Energy Conservation in Mobile Networks

Abstract

Energy conservation in mobile networks has always been concerned by the operators, as high energy consumption not only brings high cost, but also contributes to greenhouse gas emission. Besides saving energy, operators need to maintain the Quality of Service (QoS) in the network to ensure the users can enjoy satisfactory services, that is, they need to balance the QoS and energy efficiency.

In this project, I focused on finding the QoS performance and energy tradeoff in a simulated network. The network was firstly simulated using Markov chain method, and a 49-cell wrapped around model was used as the simplification of real network. Then, I developed an energy conservation scheme with basic ideal of partially shutting down some of the base stations (BSs) and turning them into sleep mode. The system would shut down the BSs according to several shutting down patterns designed by myself. Next, six ideal patterns were chosen using Pareto Optimality, and Pattern Change Thresholds were then obtained. With the thresholds, a dynamic shutting down scheme was realized, which can change the shutting down pattern based on the call arrival rate in the network, to save more energy while maintaining the QoS in the network.

After that, I did a case study to examine the capability of the designed dynamic and fixed schemes. In this case, the system call arrival rate would change according to two traffic profiles, and the QoS and energy saving results of both schemes were then compared and discussed for selection of better scheme.