Energy-Aware Offloading and Power Optimization in Full-Duplex Mobile Edge Computing-Enabled Cellular IoT Networks

Abstract

The incorporation of full-duplex (FD) communication into task offloading offers significant advantages for mobile edge computing (MEC). In this article, we investigate joint optimization of multiuser computation offloading, transmitting power, time slot, and computation resources in FD MEC-aided cellular Internet of Things (IoT) networks. For the sake of minimizing the energy consumption of the IoT nodes, we develop mixed-integer nonlinear programming with coupling constraints. To handle the coupling constraints, we derive the decoupling conditions and simplify the formulated problem into the decoupling and coupling cases. For the decoupling case, we convert the original problem equivalently into several independent subproblems. Then, for each subproblem, the transmitting power and computation offloading are optimized efficiently by the optimal condition of the objective function. After that, we derive the optimal time slot allocation by Karush–Kuhn–Tucker conditions. For the coupling case, we formulate the feasibility problem of the obtained results from the decoupling case as a two-level knapsack problem. Based on that, we develop an efficient algorithm to modify the solution of decoupling case so that its feasibility is guaranteed for the coupling constraints. Extensive simulations indicate that our proposed algorithm can effectively reduce the total energy consumption of the IoT nodes and fully improve the efficiency of MEC with respect to existing algorithms.