GTMPC: A Secure Multi-Party Computation Scheme for IIoT Data Based on Game Theory

Data sharing in the Industrial Internet of Things (IIoT) can increase productivity and reduce costs, but it is critical to ensure data security while sharing data. Secure multi-party computation (MPC) can address security problems such as data leakage during computation. However, it still faces the risk of malicious behavior from internal participants and the problem of sharing data decision-making. To address these problems, we propose a Game Theory-based secure Multi-Party Computation scheme for IIoT Data (GTMPC). Firstly, we design a traceable, verifiable and auditable industrial data sharing and computing framework for the IIoT based on blockchain. Then, we propose an MPC protocol that can resist malicious behavior from participants. It obtains the voting results through voting identifiers and multi-party multiplication. More importantly, we establish the revenue optimization models of data sharing between consumers and computing devices. Based on the above models, we model the MPC interaction-sharing process between consumers and computing devices as a Stackelberg game, and use the backward induction method to prove that there is a unique Stackelberg Equilibrium in the game. Furthermore, we also propose an MPC-solving algorithm based on the Steepest Descent Method (MPC_SDM), and obtain the optimal industrial data sharing scheme and calculation results according to the constantly updated credit score. Finally, the simulation experiment results show that the MPC_SDM algorithm is efficient and feasible in IIoT, better than the comparison algorithm, and the revenue can be increased by 13.2%.