Energy- and Spectral-Efficient Resource Allocation Algorithm for Heterogeneous Networks

TL;DR: It is shown that, in a particular setting, energy efficiency increase can be obtained in a multicell heterogeneous wireless network by sacrificing spectral efficiency, as well as a Pareto-optimal solution for energy efficiency and spectral efficiency.

Abstract: In this paper, the tradeoff between energy efficiency and spectral efficiency in multicell heterogeneous networks is investigated. Our objective is to maximize both energy efficiency and spectral efficiency of the network, while satisfying the minimum rate requirements of the users. We define our objective function as the weighted summation of energy efficiency and spectral efficiency functions. The fractional frequency reuse (FFR) scheme is employed to suppress intercell interference. We formulate the problem as cell-center boundary selection for FFR, frequency assignment to users, and power allocation. The optimal solution of this problem requires exhaustive search over all cell-center radii, frequency assignments, and power levels. We propose a three-stage algorithm and apply it consecutively until convergence. First, we select the cell-center radius for the FFR method. Second, we assign the frequency resources to users to satisfy their rate requirements and also maximize the objective function. Third, we solve the power allocation subproblem by using the Levenberg–Marquardt method. Minimum rate requirements of users are also included in the solution by using dual decomposition techniques. Our numerical results show a Pareto-optimal solution for energy efficiency and spectral efficiency. We present energy efficiency, spectral efficiency, outage probability, and average transmit power results for different minimum rate constraints. Among other results, we show that, in a particular setting, ${\text{13}}\%$ energy efficiency increase can be obtained in a multicell heterogeneous wireless network by sacrificing ${\text{7}}\%$ spectral efficiency.