Wireless-Powered OFDMA-MEC Networks With Hybrid Active–Passive Communications

Abstract

In this article, we propose a novel system model for a wireless powered mobile edge computing (MEC) network, where the Internet-of-Things (IoT) nodes perform partial offloading to the MEC server via hybrid backscatter communication (BackCom) and active radio (AR) following an OFDMA protocol, and maximize the system computation bits (SCB). For the case of the system having more subchannels than IoT nodes, we formulate the SCB maximization problem that requires the joint optimisation of the transmit power and time, subchannel allocation, computation frequency and time of the MEC server, as well as the IoT nodes’ BackCom time and reflection coefficients, transmit power and time for AR-based offloading, local computing time and frequencies, subject to the MEC server’s computation capacity and the quality-of-service (QoS) and energy-causality constraints of each IoT node. By applying the proof by contradiction and time sharing relaxation, we transform the formulated problem into a convex one and then solve it by using the existing convex tools. For the case of the system having less subchannels than IoT nodes, we propose a dynamic subchannel allocation scheme that allows each IoT node to choose one task-offloading mode from three modes: HAPR, BackCom only, and AR only, while ensuring that no more than one IoT node occupies a subchannel at any time. The SCB are maximized by first determining the subchannel allocation and mode selection of each IoT node and then optimising the remaining resource allocation for the MEC server and all IoT nodes under the obtained subchannel assignment and mode selection. Simulations validate the superior performance of the proposed schemes over several benchmark schemes from the SCB perspective.