Joint Optimization of Task Assignment and Power Allocation for NOMA-Aided MEC Systems

Abstract

In this paper, task assignment and power allocation are investigated for the the non-orthogonal multiple access (NOMA)-aided mobile edge computing (MEC) system. Based on the different channel conditions and central processing units (CPUs), users can offload computational tasks to the MEC server or process tasks locally. In order to minimize the energy consumption of the proposed NOMA- aided MEC system, the task assignment and power allocation optimization problem is formulated. Based on the insight derived from the delay constraint, the closed-form expressions of task ratios and transmit power are derived. Furthermore, the optimality of the derived closed-form solutions is analyzed, which shows that the optimal task assignment of any user is based on the energy consumption efficiency (ECE) of offloading and local computing. Simulation results indicate that: i) the derived closed-form solutions can significantly reduce the energy consumption of the NOMA-MEC system, and ii) the analysis of the derived closed-form solutions is confirmed.