Energy efficiency priority admission control and service transfer algorithm based on Markov decision processes in hierarchical networks

In a hierarchical cellular network environment, both macrocell and femtocell are deployed to improve network performance. The admission control problem for hierarchical network was addressed when all types of handovers were considered. The admission control and service transfer problem was formulated as a Markov decision process (MDP) optimization problem with constraints on the blocking probability and dropping probability and the objective of maximizing the average system energy efficiency reward function. When there are idle resources in a femtocell, the services connected to a macrocell with users located in the coverage of the femtocell could be chosen to be handed over to the femtocell. The optimal policy is obtained by solving a linear programming problem. Simulation results show that the proposed energy efficiency priority admission control and transfer strategy effectively improves the resource utilization of the femtocell with the maximum reward, and decreases blocking probability and dropping probability. Meanwhile, reducing resource utilization of the macrocell will make it possible to provide services to users within a larger coverage area.