Bandwidth Allocation For Wireless Multimedia Social Network In Cloud Environments

Distributed resource allocation is critical for efficient sharing multimedia contents in cloud-based wireless multimedia social network environments.As wireless communication technologies evolved,the mobile devices such as smartphone,panel computer and personal digital assistant have experienced an explosive growth in recent years.Therefore,how to efficiently allocate resources has become a key challenge in managing the multimedia service networks.First,a cloud-based multimedia service system architecture is proposed to overcome the limited bandwidth allocation problem in the context of social network,in which bandwidth limited mobile users are allowed to directly acquire live multimedia streaming from the desktop users rather than the cloud based on their social relationships.Then,a theoretical framework is presented for bandwidth allocation from desktop users to mobile users by a dynamic resource pricing process in the proposed bandwidth management system,where price-sensitive users and QoS-sensitive users are considered,and all users target at maximizing their total utilities.Finally,an iterative allocation algorithm is designed to simulate the bandwidth allocation process with respect to shared bandwidth and price.Simulation results verify the effectiveness of our proposed pricing model and allocation algorithm in terms of convergence and efficiency.The thesis studies the bandwidth allocation and pricing in the multimedia social network and opens the application of the iterative algorithm and dynamic pricing,which can provide the reference value for further researches on multimedia social network.