Towards heterogeneous network convergence: policies and middleware architecture for efficient flow assignment, rate allocation and rate control for multimedia applications

  • Authors:
  • Jatinder Pal Singh;Tansu Alpcan;Xiaoqing Zhu;Piyush Agrawal

  • Affiliations:
  • Deutsche Telekom AG, Berlin, Germany;Deutsche Telekom AG, Berlin, Germany;Stanford University, Stanford, CA;Indian Institute of Technology, Kanpur, India

  • Venue:
  • Proceedings of the 2007 Workshop on Middleware for next-generation converged networks and applications
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The convergence in access networks brings forth the challenge of resource allocation for applications with varying utilities over networks with time-varying and heterogeneous characteristics. Owing to an ever increasing demand and popularity, multimedia applications have been a key driving force behind the efforts targeted towards architectures, industrial standards, and polices for convergence. Multimedia applications use protocols and transport dynamics that constrain the manner in which these applications can utilize access networks. In this work, we propose middleware architecture and policies that target optimal rate allocation for multimedia applications by monitoring and stochastically modeling the variation in characteristics of the access networks, using the utilities of the media application when known, and accounting for the constraints imposed by application specific transport mechanisms. The proposed measures leverage our optimal rate allocation, rate control, and interface assignment techniques [17, 6, 24] to achieve enhanced multimedia performance and efficient utilization of the access networks.