TY - JOUR
T1 - Feed Forward Bandwidth Indication (FFBI)
T2 - Cooperation for an accurate bandwidth forecast
AU - Haddad, Rami
AU - McGarry, Michael P.
N1 - Video bandwidth forecasts can empower video transport mechanisms with a new intelligence that can increase the efficiency of Dynamic Bandwidth Allocation. We exploit the fact that for pre-recorded video, the size of every video frame is known prior to the video being delivered.
PY - 2012/3/15
Y1 - 2012/3/15
N2 - Video bandwidth forecasts can empower video transport mechanisms with a new intelligence that can increase the efficiency of Dynamic Bandwidth Allocation. We exploit the fact that for pre-recorded video, the size of every video frame is known prior to the video being delivered. We propose Feed-Forward Bandwidth Indication (FFBI) which feeds video frame sizes forward in a sequence of video frames. We extend FFBI to live video by introducing a delay at the source equivalent to the forecast window. We compare FFBI to the most accurate forecast methods found in the literature. With network transport of video projected to supplant other transport mechanisms over the next few years, we conduct a performance analysis of FFBI within Ethernet Passive Optical Networks (EPONs). We find that the use of FFBI can provide a 50% reduction in queueing delay compared to the use of no forecasting and a 35% reduction in queueing delay compared to other forecasting methods. In addition, we find that FFBI can provide a very significant reduction in queueing delay variation compared to the use of no forecasting or other forecasting methods.
AB - Video bandwidth forecasts can empower video transport mechanisms with a new intelligence that can increase the efficiency of Dynamic Bandwidth Allocation. We exploit the fact that for pre-recorded video, the size of every video frame is known prior to the video being delivered. We propose Feed-Forward Bandwidth Indication (FFBI) which feeds video frame sizes forward in a sequence of video frames. We extend FFBI to live video by introducing a delay at the source equivalent to the forecast window. We compare FFBI to the most accurate forecast methods found in the literature. With network transport of video projected to supplant other transport mechanisms over the next few years, we conduct a performance analysis of FFBI within Ethernet Passive Optical Networks (EPONs). We find that the use of FFBI can provide a 50% reduction in queueing delay compared to the use of no forecasting and a 35% reduction in queueing delay compared to other forecasting methods. In addition, we find that FFBI can provide a very significant reduction in queueing delay variation compared to the use of no forecasting or other forecasting methods.
KW - Bandwidth forecasting
KW - Bandwidth prediction
KW - Multimedia
KW - Video
UR - https://scholarworks.utep.edu/elec_comp_papers/44
UR - https://www.sciencedirect.com/science/article/pii/S0140366412000059
UR - http://www.scopus.com/inward/record.url?scp=84857788409&partnerID=8YFLogxK
U2 - 10.1016/j.comcom.2012.01.004
DO - 10.1016/j.comcom.2012.01.004
M3 - Article
SN - 0140-3664
VL - 35
SP - 748
EP - 758
JO - Computer Communications
JF - Computer Communications
IS - 6
ER -