Resource Allocation for D2D Video Multicast Using Multi-Leader Multi-Follower Stackelberg Game With Personalized Incentives

With the growing demand of mobile multimedia services, device-to-device (D2D) multicast provides an efficacious solution to local content sharing. The radio resource needs to be allocated efficiently to enhance the content sharing via D2D multicast. However, the existing resource allocation schemes for solving D2D multicast usually ignore the attribution problem of mobile users to multiple communities and the problem of social differentiation. Unlike the existing works, this paper performs the D2D video multicast in two steps: community formation and resource allocation. In the process of D2D multicast community formation, the best core users for content distribution in the network are selected in consideration of both physical and social factors, and detect whether there is overlap in the communities to which others belong. Then, we explore mobile users' social features including the historical request file similarity, the QoS request differentiation and the random mobility characteristic to arrange the best attribution schemes targeting at not only ensuring the quality of the video multicast service but also maximizing the content hit ratio. In the resource allocation step, a multi-agent hierarchical learning (MAHL) algorithm with personalized incentives based on multi-leader multi-follower Stackelberg game is proposed to maximize the throughput of the network. Simulations are conducted to reveal that compared with the other three benchmark algorithms, the proposed algorithm can significantly improve the throughput of the network under different scenarios.