Energy optimization in cellular networks with micro-/pico-cells using Markov Decision Process

Power conservation of communication networks at the system level has received much attention due to the tremendous increase in the number of infrastructure equipment in recent years. In this paper, we propose a teletraffic-based optimization approach to minimize the power consumption of cellular networks with three micro- or pico-cells. When traffic load in the middle cell is sufficiently low, the cell can be switched off and its users will be covered by increasing the transmission power of one sector in each of the two neighboring cells. A Finite Markov Decision Process (FMDP) is constructed and analyzed in order to minimize network transmission power consumption. Numerical results demonstrate that minimized power consumption can be achieved using our FMDP model. The long-term power consumption can also be predicted.