Collaborative content dissemination based on game theory in multimedia cloud

Abstract

With the development of multimedia clouding applications and broadband technology, multimedia group communication applications, such as multi-point video conference, major news and events live, internet-on-demand TV, network video monitoring and so on, continue to emerge. However, limited network bandwidth, burst of incoming service requests, and imbalance of content dissemination seem to be the most important factors that affect users’ experience. To this end, this paper proposes a collaborative content dissemination method based on game theory in multimedia cloud. In order to obtain the optimal number of service user in each group, integrated utility value of each user is calculated, and the improved particle swarm optimization algorithm is used. Moreover, the matching problem between service users and non-service users is formulated into game theory. In order to avoid the blindness and selfishness of non-service users to choose the service users, an evolutionary game is aimed at achieving the stability by learning the connection from the non-service users. A Stackelberg game is aimed at obtaining optimal pricing policy and bandwidth allocation of service users. The numerous experiment results show that the proposed method can effectively improve users’ experience of those who have weak system capabilities, reduce the average time of the content dissemination and decrease the total cost of the multimedia cloud user.