On Exploiting Bandwidth Predictability to Optimize IP/DTV Resource Allocation in DVB-S2 Networks

Convergence of digital video broadcasting networks and IP networks represents undoubtedly the main evolution in next generation of services combining one-way digital television and interactive Internet applications. Within this mixed context, an important issue arises concerning the management of resources between IP applications and DTV services. The paper investigates this issue and presents a solution based on an interactive dynamic bandwidth management system (iDBMS), with prediction functionalities and using adaptive filtering techniques for optimal provisioning of concurrent IP and Digital TV services over a fast DVB-S2 network. We compare the performances of the iDBMS solution with existing schemes using the opportunistic data insertion mode (ODI) under OPNET simulator. We show that the proposed mechanism allows maximizing the bandwidth allocation, while sharing in real-time the available bandwidth between digital TV and IP services.