Task Offloading and Resource Allocation in D2D-Assisted Mobile Edge Computing Based on Game Theory

The task processing capabilities of mobile edge computing (MEC) networks can be significantly enhanced through the integration of device-to-device (D2D) technology, rendering them particularly suitable for computation-intensive applications in the Internet of Things (IoT). However, the opportunistic behaviors of computation-intensive users (CUs) will reduce the fairness of resource allocation, which will increase the task offloading delay for others inevitably. To cope with this issue, a three-layer collaborative offloading scheme including edge offloading, D2D offloading, and cloud offloading is proposed to provide abundant computing resources, thus improving the fairness. Built on this scheme, the offloading mode selection, channel selection, offloading side selection, offloading ratio allocation, and computing resource allocation are jointly optimized to minimize the task processing delay while ensuring high fairness. Such an optimization problem is a mixed-integer nonlinear programming (MINLP). Therefore, an evaluation-based distributed task offloading strategy selection (E-DTOSS) algorithm is designed, where the task offloading problem can be decomposed into three sub-problems. The offloading strategy selection problem, which is NP-hard, is proven to be an exact potential game and its pure-strategy Nash equilibrium point can be obtained by the E-DTOSS algorithm. The offloading ratio allocation problem and computing resources allocation problem are solved by dual ascent method. Simulation results verify that the E-DTOSS algorithm can converge to the Nash equilibrium point while ensuring high fairness. As a result, the task processing delay can be reduced compared with the benchmark schemes.