Edge computing-empowered task offloading in PLC-wireless integrated network based on matching with quota

Abstract

The integration of the fifth-generation (5G) wireless communication with power line communication (PLC) provides an efficient solution to accommodate the connection demands of massive industrial internet of things (IIoT) devices. Moreover, edge computing can provide sufficient computing services for IIoT devices. By empowering PLC-wireless integrated networks with edge computing, devices can offload task data to adjacent edge servers through either PLC or 5G subchannels by leveraging orthogonal frequency division multiplexing (OFDM). However, task offloading in the edge computing-empowered PLC-wireless integrated network faces several challenges such as channel access conflict, dynamic network state, as well as long-term queue stability assurance. In this article, we construct a one-to-many task offloading architecture to maximize the total throughput. Lyapunov optimization is adopted to convert the NP-hard task offloading problem. Then, a pricing-based matching with quota task offloading algorithm is proposed to handle channel access conflicts among IIoT devices. Simulation evaluation demonstrates that the proposed algorithm realizes excellent performances in throughput, energy consumption, and queuing backlog.