Macrocell-Queue-Stabilization-Based Power Control of Femtocell Networks

We study the power control for femtocell networks in a heterogeneous spectrum-sharing network consisting of macrocells and femtocells. We propose a new approach that utilizes macrocell queue state information for femtocell power control. We formulate the underlying stochastic optimization problem to maximize average femtocell throughput under the constraint of macrocell queue stability by using the macrocell queue length information as a Lagrange weight of macrocell throughput. This formulation considers both cross-tier interference between femtocells and macrocells and intra-tier interference among femtocells. We design a centralized algorithm and a distributed algorithm for femtocells to decide downlink power for each sub-channel dynamically. Our simulation tests show that using local information, the distributed algorithm implementation can also achieve similar average femtocell throughput as the centralized algorithm by paying a penalty of larger macro queue length variation.