3rd International ICST Conference on Mobile Multimedia Communications

Research Article

A Bandwidth Dependent Smoothing Algorithm for Interactive Video Streaming in UMTS Systems

Download460 downloads
  • @INPROCEEDINGS{10.4108/ICST.MOBIMEDIA2007.1691,
        author={Pietro Camarda and Domenico Striccoli and Mariella Ragno},
        title={A Bandwidth Dependent Smoothing Algorithm for Interactive Video Streaming in UMTS Systems},
        proceedings={3rd International ICST Conference on Mobile Multimedia Communications},
        proceedings_a={MOBIMEDIA},
        year={2010},
        month={5},
        keywords={Video Streaming UMTS Smoothing Interactivity Available Bandwidth.},
        doi={10.4108/ICST.MOBIMEDIA2007.1691}
    }
    
  • Pietro Camarda
    Domenico Striccoli
    Mariella Ragno
    Year: 2010
    A Bandwidth Dependent Smoothing Algorithm for Interactive Video Streaming in UMTS Systems
    MOBIMEDIA
    ICST
    DOI: 10.4108/ICST.MOBIMEDIA2007.1691
Pietro Camarda1,*, Domenico Striccoli2,*, Mariella Ragno2,*
  • 1: Politecnico di Bari Via E. Orabona, 4 70125 – BARI (ITALY) +39 080 5963642
  • 2: Politecnico di Bari Via E. Orabona, 4 70125 – BARI (ITALY) +39 080 5963301
*Contact email: camarda@poliba.it, d.striccoli@poliba.it, mariellaragno@yahoo.it

Abstract

In the recent past, a growing number of services, such VBR video transmission, have been implemented in UMTS cellular systems. In such a context, to reduce the high bit rate variability of VBR streams, several smoothing techniques, performed at server side, have been developed. They regularize the bit rate of transmitted data maintaining, at the same time, a constant video quality at receiving side. In this paper, a novel smoothing algorithm, the Buffer Dependent Smoothing Algorithm (BDSA), has been developed and analyzed. It schedules VBR video data, taking into account both the feedback information on the real residual free buffer size coming from the client terminal, and the available bandwidth information. Numerical results show the BDSA effectiveness, in terms of losses, if compared with the classical smoothing algorithms known by literature.