Abstract
The study of spectral efficiency (SE) in cellular networks has attracted significant research interest in the wireless community. Heterogeneous networks (HetNets) have emerged as a new paradigm to meet the fast growing demands for higher capacity and coverage in cellular networks. Previous works have focused on investigating either the power and bandwidth allocation problem or the power allocation problem for SE maximization in HetNets. These works were carried out under the assumption that users have already been associated. This paper presents a simple and effective method to optimize the SE of two-tier HetNets. Specifically, the joint optimization of user association and power control is formulated as a mixed integer programming problem. To tackle the non-convexity of the optimization problem, the Lagrange duality theory is applied to decompose the original problem into a user association sub- problem and a power control sub-problem, which are solved alternatively. Simulation results are used to demonstrate the fast convergence rate and the noticeable performance gains of the proposed algorithm as compared with conventional user association schemes under the assumption of equal power allocation.