A Novel Multi-Channel Data Broadcast Scheme for Multimedia Database Systems

MultiMedia DataBase Management System (MMDBMS) becomes more popular in recent years, which supports complex and large multimedia data like images, audios, and videos etc. Data Broadcasting is an attractive approach for data dissemination to improve the limitations in mobile environment, such as narrow bandwidth, unreliable connections, and battery limitation. However, existing data broadcast schemes are inefficient for MMDBMS. In this paper, we present four novel multimedia data broadcast schemes (namely, SDAA, MDAA, AEA, and COA) specifically for wireless multichannel communications. The major strategies are scalable coding to generate data segments to different qualities, indexing and channel assignment to minimize the expected waiting time for clients. We prove theoretically that SDAA is a 2-approximation. COA performs best when we release the constraints and it can be judged as an theoretical lower bound, while AEA outputs local optimal solution with quality allocation constraints. Finally, SDAA+AEA form a best scheduling for practical applications. We also provide numerical experiments to evaluate the system performance, proving the efficiency of our schemes.