Utility based Energy-efficient Resource Allocation Algorithm in OFDM System

Resource allocation in Orthogonal Frequency Division Multiplexing (OFDM) system with mixed traffic is studied in this paper. The sub-carriers and power of a base station (BS) are allocated to users to maximize energy efficiency, which is measured by utility per Watt. We model the allocation of sub-carriers and power as a mixed integer nonlinear programming (MINLP) problem, which is computationally forbidden due to its non-concave objective function, non-convex feasible region and integer variables. To simplify the solution of the problem, a two-step resource allocation (TSRA) algorithm is proposed. In the first step, the integer variables are transformed into continuous variables and uniform power allocation is assumed. On this basis, sub-carrier allocation is obtained using KKT condition. Knowing the allocation of sub-carrier, the optimal power allocation is solved by an iterative sub-gradient method in the second step. Simulation results show that our algorithm can provide 25% energy efficiency improvement compared to conventional algorithms.