Summary: | This paper considers a load-aware energy efficiency maximization problem for downlink heterogeneous cellular networks with attention to unequal user priorities, load balance, quality of service (QoS) requirements, and power control. The load-aware energy efficiency of the networks is defined as the ratio of the effective rate and power consumption and mathematically formulated as a mixed-integer optimization problem in fractional forms with multi-constraints. A two-layer iterative algorithm with low computational complexity is proposed to address this optimization problem. In the outer layer, the objective function is converted into an equivalent problem in subtractive form by using the Dinkelbach method. The joint resource allocation involves cell-user association and power control that are decoupled over the independent portions of the system. In the inner layer, we search the association indices and transmit power parameters via solving a class of convex optimization problems governed by constraints. Specifically, by introducing some new auxiliary variables and using the Lagrangian dual method, the closed-form optimal association strategy and power control solution are obtained. The numerical simulations show that the proposed algorithm is guaranteed to converge and gives rise to higher energy efficiency than the existing one.
|