Optimal and Fair Energy Efficient Resource Allocation for Energy Harvesting-Enabled-PD-NOMA-Based HetNets

In this paper, the tradeoff among the energy efficiency, fairness, harvested energy, and system sum rate is studied. In this regard, various fairness methods, namely, max-min fairness, proportional fairness, and minimum delay potential fairness in power-domain non-orthogonal multiple access-based heterogeneous cellular networks are investigated. In order to perform successive interference cancellation (SIC), we use two ordering approaches and compare their performance. To this end, we propose joint subcarrier and power allocation algorithms to achieve fair energy efficient resource allocation for each fairness method and SIC ordering. Since the proposed optimization problems are non-convex and intractable, the existing methods to solve the convex problems could not be directly used. To overcome this difficulty, an iterative algorithm based on successive convex approximation is used. Moreover, to show the optimality gap of the proposed solution method, an optimal approach based on the monotonic optimization is applied in which we first transform each of the proposed optimization problems into a monotonic optimization problem of canonical form, and then, we obtain the optimal solution of each problem, which coincides with the optimal solution of the original non-convex problem. We finally study the performance of the proposed schemes using simulations for different values of the system parameters.

I. Introduction

With increasing the demand of high data rate cellular services and due to the limitations on the available radio resources, there are many challenges which are expected to be addressed in the next generation of cellular networks. Statistical data show that the required traffic of mobile becomes approximately doubling every year [1], and also, they show that wireless cellular systems occupy 0.5% of the total energy consumption in the world [2]. Consequently, applying new schemes in resource allocation is very important to achieve appropriate energy efficiency (EE) and spectral efficiency (SE).