Multi-Objective Energy Efficient Resource Allocation and User Association for In-Band Full Duplex Small-Cells

Abstract

In this paper, we develop a framework to maximize the network energy efficiency (EE) by optimizing joint userbase station (BS) association, subchannel assignment, and power control considering an in-band full-duplex (IBFD)-enabled smallcell network. We maximize EE (ratio of network aggregate throughput and power consumption) while guaranteeing a minimum data rate requirement in both the uplink and downlink. The considered problem belongs to the category of mixed-integer non-linear programming problem (MINLP), and thus is NPhard. To cope up with this complexity and to derive a tradeoff between system throughput and energy utilization, we first restate the considered problem as a multi-objective optimization problem (MOOP) aiming at maximizing system's throughput and minimizing system's energy consumption, simultaneously. This MOOP is then tackled by using-constraint method. To do so, we first transform the binary subchannel and BS assignment variables into continuous ones without altering the feasible region of the problem and then approximate the nonconvex rate functions through majorization-minimization (MM) approach. Simulation results are presented to demonstrate the effectiveness of our proposed algorithm in improving network's EE compared to the existing literature. Furthermore, simulation results unveil that by employing the IBFD capability in OFDMA networks, our proposed resource allocation algorithm achieves a 69% improvement in the EE as compared to the half-duplex system for practical values of residual self-interference.

GA based energy efficiency resource allocation in heterogeneous networks with MRC

“…Accordingly, the EE optimization problem for the downlink OFDMA HetNets can be written as max EE, (8) subject to: C1 (Orthogonality constraint):…”

“…The scalarization methods convert the MOOP into a series of the parametric single‐objective optimization problem (SOOP), whilst vectorization methods solve the MOOP directly. In [8], the EE optimization problem was first transformed into MOOP form, and then the MOOP was solved by using the $$\epsilon$$‐constraint method. For the multiple objective functions, one objective function is selected as the main objective function, and all other objective functions are used as constraint conditions.…”

Energy Efficiency Maximization in the Uplink Delta-OMA Networks

“…To the best of our knowledge, the joint optimization of SR and sum power (SP) in terms of finding optimal value of δ, p, and ρ have not been investigated before where we will discuss it in the next subsequent sections. Note that ρ, p and δ are the vector representation of the variables It can be easily proved that a problem with the objective of maximizing EE (which is a ratio of total rate to the power consumption) is equivalent to a multi-objective optimization which the objectives are maximizing total rate (the nominator of the EE) and minimizing total power consumption (the denominator of the EE) [12]. Therefore, invoking this property, we formulate a multi-objective optimization problem in the next section.…”

Energy Efficiency Maximization in the Uplink Delta-OMA Networks

“…To the best of our knowledge, the joint optimization of SR and sum power (SP) in terms of finding optimal value of δ, p, and ρ have not been investigated before where we will discuss it in the next subsequent sections. Note that ρ, p and δ are the vector representation of the variables It can be easily proved that a problem with the objective of maximizing EE (which is a ratio of total rate to the power consumption) is equivalent to a multi-objective optimization which the objectives are maximizing total rate (the nominator of the EE) and minimizing total power consumption (the denominator of the EE) [10]. Therefore, invoking this property, we formulate a multi-objective optimization problem in the next section.…”