QoS-driven user association for load balancing and interference management in HetNets

In this paper, we investigate how to jointly optimize the Almost Blank Subframe (ABS) ratio computation and the user association in Heterogeneous Networks (HetNets) while considering both the load of each BS and the users' QoS requirements. Our problem is a combinatorial NP-hard one. To solve it, we propose a centralized heuristic algorithm that aims to discover the optimal ABS ratio while maximizing the number of served users in the case of a two-tier HetNet. The association is based on the average received SINRs in both protected and non-protected subbands and on the actual loads in the cells. Numerical results show the efficiency of our proposed scheme in enhancing the system admission rate and balancing the load.