SGCS: An Intelligent Stackelberg-Game-Based Computation Offloading and Resource Pricing Scheme in Blockchain-Enabled MEC for IIoT
Bing Lin, Xuzhan Chen, Xing Chen, Yun Ma, Naixue Xiong
Abstract
Mobile Edge Computing (MEC) offers low-latency and flexible computing services for mobile devices (MDs) in industrial Internet-of-Things (IIoT). Edge servers (ESs) in general belong to different subjects and will focus on their own interests. They may be reluctant to provide computation resources to MDs without appropriate incentives. Meanwhile, there is a trust issue in trading computation resources between ESs and MDs. Due to the complex interaction between ESs and MDs, it is a challenge for ESs to gain satisfactory revenue through reasonable resource pricing strategies, and for MDs to improve their quality of experience (QoE) through efficient computation offloading strategies. This paper proposes a Stackelberg Game-based Computation offloading and resource pricing Scheme (SGCS) in blockchain-enable MEC for IIoT. Firstly, a blockchain-based resource trading framework is designed to enable trusted resource transactions. Secondly, a multi-leader multi-follower Stackelberg game is presented to analyse the complex interactions in the multi-ES and multi-MD environments. Finally, the Iterative Proximal Algorithm (IPA) for MDs’ offloading decision and the Sub-gradient based Iterative Pricing Algorithm (SIPA) for ESs’ pricing decision are proposed respectively, which guarantees that the game converges to a Stackelberg Equilibrium (SE). Compared with MADDPG, GA and PSO-GA (i.e., benchmark strategies), the average disutility of MDs with our proposed scheme is reduced by 6.95%-13.07%, 3.09%-20.41%, and 2.36-17.22%, respectively. Moreover, with the increase of the number of MDs, our proposed scheme has better robustness, which can effectively deal with large-scale scenarios.