Resource allocation in heterogeneous networks with eICIC

Joint optimization of user association, resource allocation and Almost Blank Subframe (ABSF) based enhanced Inter-Cell Interference Coordination (eICIC), on which macro-cells mute for scheduling the pico users, is of great importance to further improve the network performance in HetNets, however, the introduce of ABSF makes analyzing resource allocation extremely difficult because this results in different interference patterns. In this paper, we proposed a novel approach to determine the association of users, allocation of resources and the proportion of ABSFs. We formulate a system throughput maximization problem with proportional rate constraint, compared with using log-utility, which can not be utilized to adjust resource allocation, our model can help us analyze and explicitly control the resource allocation among users under different interference patterns. The difficulty of our formulated problem comes from that it is NP-hard, to overcome this challenge, we relax the integer constraint and the original problem is decomposed into two subproblems. Firstly, for a given ABSFs proportion, we obtain an insight that the optimal user association is determined by the achievable rate ratio offered by macrocell and picocell. Also, for a given association of users, we figure out the explicit expressions for the optimal resource allocation and ABSFs proportion. By exploiting the characterization of the optimal solution, we proposed a low complicity algorithm. Simulation results show the effectiveness of the proposed algorithm.