MotiShare: Incentive Mechanisms for Content Providers in Heterogeneous Time-varying Edge Content Market
Quyuan Wang,Songtao Guo,Jiadi Liu,Chengsheng Pan,Li Yang
DOI: https://doi.org/10.1109/tsc.2021.3124419
IF: 11.019
2021-01-01
IEEE Transactions on Services Computing
Abstract:With the development of edge computing and sharing economy, more services and contents are decentralized to the edge of the network. At present, most existing studies combine the content caching with service offloading mechanisms from the perspective of edge services to optimize network performance. However, few studies focus on what strategies the content providers (CPs) can implement to maximize their utilities. In order to motivate content providers to be more willing to share their contents, it is necessary to study the incentive mechanisms in the edge content market so that content audiences (CAs) can pay reasonable prices for high-quality contents. In this article, we first characterize the content supply and demand model, the CPs’ cost and utilities in edge content market by considering both the time sensitivity of edge content and the heterogeneity of CAs. Furthermore, according to the edge content market environment characteristics, we divide the edge content market into a monopoly environment, where the content is only provided by a certain CP, and an open environment, where content services are provided by multiple CPs. In the monopoly environment, we establish a two-stage Stackelberg game to design the incentive mechanism. In the open environment, also we formulate the competitive behavior among CPs as a stochastic game. Since the CPs are not aware of each other's strategies and environmental uncertainty, the reinforcement learning-based algorithm (RLIMO) is used to derive the pricing strategy of CP. Finally, numerical results show that the proposed incentive mechanisms are reliable and effective.
computer science, information systems, software engineering