a comparison of layering and stream replication video multicast schemes

46
A Comparison of Layering and Stream Replication Video Multicast Schemes Taehyun Kim and Mostafa H. Ammar Networking and Telecommunications Group Georgia Institute of Technology Atlanta, Georgia

Upload: talli

Post on 16-Jan-2016

34 views

Category:

Documents


0 download

DESCRIPTION

A Comparison of Layering and Stream Replication Video Multicast Schemes. Taehyun Kim and Mostafa H. Ammar Networking and Telecommunications Group Georgia Institute of Technology Atlanta, Georgia. Research Goal. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: A Comparison of Layering and Stream Replication Video Multicast Schemes

A Comparison of Layering and Stream Replication Video

Multicast Schemes

Taehyun Kim and Mostafa H. AmmarNetworking and Telecommunications Group

Georgia Institute of TechnologyAtlanta, Georgia

Page 2: A Comparison of Layering and Stream Replication Video Multicast Schemes

Research Goal

A systematic comparison of video multicasting schemes designed to deal with heterogeneous receivers Replicated streams Cumulative layering Non-cumulative layering

Page 3: A Comparison of Layering and Stream Replication Video Multicast Schemes

Stream Replication

Multiple video streamsSame content with different data ratesReceiver subscribes to only one streamExample

DSG (Cheung, Ammar, and Li, 1996) SureStream of RealNetworks

Intelligent streaming of Microsoft

Page 4: A Comparison of Layering and Stream Replication Video Multicast Schemes

Replicated Stream Multicast

Page 5: A Comparison of Layering and Stream Replication Video Multicast Schemes

Cumulative Layering

1 base layer + enhancement layersBase layer

Independently decoded

Enhancement layer Decoded with lower layers Improve the video quality

Example RLM (McCanne, Jacobson, Vetterli, 1996) LVMR (Li, Paul, and Ammar, 1998) MPEG-2/4, H.263 scalability modes

Page 6: A Comparison of Layering and Stream Replication Video Multicast Schemes

Layered Video Multicast

Page 7: A Comparison of Layering and Stream Replication Video Multicast Schemes

Layering or Replication?

Common wisdom states: “Layering is better than replication” But it depends on

Layering bandwidth penalty Specifics of encoding Protocol complexity Topological placement of receivers

Page 8: A Comparison of Layering and Stream Replication Video Multicast Schemes

Bandwidth Penalty

Information theoretic results R(P, 2) R(P, 1, 2)

Packetization overhead Syntactically independent layering

Picture header GOP information Macroblock information

Page 9: A Comparison of Layering and Stream Replication Video Multicast Schemes

Experimental Comparison

Page 10: A Comparison of Layering and Stream Replication Video Multicast Schemes

Comparison by DP

J. Kimura, F. A. Tobagi, J. M. Pulido, P. J. Emstad, "Perceived quality and bandwidth characterization of layered MPEG-2 video encoding", Proc. of the SPIE, Boston, MA, Sept. 1999

Page 11: A Comparison of Layering and Stream Replication Video Multicast Schemes

Providing a Fair Comparison

Need to insure that each scheme is optimized

Two dimensions Selection of stream/layer rates Assignments of streams/layers to

receivers

Page 12: A Comparison of Layering and Stream Replication Video Multicast Schemes

Rate allocation

Cumulative layering Optimal receiver partitioning algorithm

(Yang, Kim, and Lam)

Stream replication Cumulative rate allocation

Page 13: A Comparison of Layering and Stream Replication Video Multicast Schemes

Stream assignment

Cumulative layering Assign as many layers as possible

Stream replication Greedy algorithm

Page 14: A Comparison of Layering and Stream Replication Video Multicast Schemes

Comparison Methodology

Model of network Topology Available bandwidth Placement of source and receivers

Determine optimal stream rates and allocation

Evaluate performance

Page 15: A Comparison of Layering and Stream Replication Video Multicast Schemes

Performance Metrics

Average reception rateTotal bandwidth usageAverage effective reception rate Efficiency

usage bandwidth Totalrate reception effective Total

Page 16: A Comparison of Layering and Stream Replication Video Multicast Schemes

Network Topology

GT-ITM Number of server = 1 Number of receivers = 1,640 Number of transit domains = 10

Number of layers = 8Amount of penalty = 25%

Page 17: A Comparison of Layering and Stream Replication Video Multicast Schemes

Data reception rate

Page 18: A Comparison of Layering and Stream Replication Video Multicast Schemes

Bandwidth usage

Page 19: A Comparison of Layering and Stream Replication Video Multicast Schemes

Effective reception rate

Page 20: A Comparison of Layering and Stream Replication Video Multicast Schemes

Efficiency

Page 21: A Comparison of Layering and Stream Replication Video Multicast Schemes

Effect of overhead

Page 22: A Comparison of Layering and Stream Replication Video Multicast Schemes

Effect of the number of layers

Page 23: A Comparison of Layering and Stream Replication Video Multicast Schemes

Clustered Distribution

Topology consideration Layering favors clustered receivers Stream replication favors randomly

distributed receiversSimulate when receivers are clustered

within one transit domain

Page 24: A Comparison of Layering and Stream Replication Video Multicast Schemes

Effective reception rate

Page 25: A Comparison of Layering and Stream Replication Video Multicast Schemes

Protocol Complexity

Layered video multicasting Multiple join for a receiver Large multicast group size

Replicated stream video multicasting One group for a receiver Small multicast group size

Page 26: A Comparison of Layering and Stream Replication Video Multicast Schemes

Average group size

Page 27: A Comparison of Layering and Stream Replication Video Multicast Schemes

Conclusion

Identified the factors affecting relative merits of layering versus replication Layering penalty Specifics of the encoding Topological placement Protocol complexity

Developed stream assignment and rate allocation algorithm

Investigated the conditions under which each scheme is superior

Page 28: A Comparison of Layering and Stream Replication Video Multicast Schemes

Optimal Quality Adaptation for MPEG-4 Fine-Grained

Scalable Video

Taehyun Kim and Mostafa H. AmmarNetworking and Telecommunications Group

Georgia Institute of TechnologyAtlanta, Georgia

Page 29: A Comparison of Layering and Stream Replication Video Multicast Schemes

Related Work (1/2)

S. Nelakuditi, et al, “Providing smoother quality layered video stream,” NOSSDAV 2000

Goals Achieving smoother quality for layered

CBR video using receiver buffer Minimizing quality variation (maximizing

runs of continuous frames)

Page 30: A Comparison of Layering and Stream Replication Video Multicast Schemes

Algorithm

Forward scan Switching between select and discard

phase Entering select phase if buffer is full Entering discard phase if buffer is empty

Backward scan Exploiting the residual buffer Extending each run

Page 31: A Comparison of Layering and Stream Replication Video Multicast Schemes

Bandwidth Model

Page 32: A Comparison of Layering and Stream Replication Video Multicast Schemes

Experimental Result

Page 33: A Comparison of Layering and Stream Replication Video Multicast Schemes

Experimental Result

Page 34: A Comparison of Layering and Stream Replication Video Multicast Schemes

Related Work (2/2)

D. Saparilla, et al, “Optimal streaming of layered video,” INFOCOM 2000

Goal Investigating the bandwidth allocation

problem to minimize loss probability Modeling the source video and the

available bandwidth by stochastic process

Page 35: A Comparison of Layering and Stream Replication Video Multicast Schemes

Main Result

Static policy Allocating bandwidth in proportion to

long run average data rate Optimal for infinite length, independent

layeringThreshold-based policy

If the base layer buffer is below a threshold, allocate bandwidth to the base layer

Page 36: A Comparison of Layering and Stream Replication Video Multicast Schemes

Research Goal of MPEG4 FGS Quality Adaptation

Maximization of the perceptual video quality by minimizing quality variation

Accommodation of the mismatch between Rate variability of VBR video Available bandwidth variability

Page 37: A Comparison of Layering and Stream Replication Video Multicast Schemes

MPEG4 FGS Hybrid Scalability

Base layerEnhancement layer

FGS layer: improving video quality FGST layer: improving temporal

resolution

Page 38: A Comparison of Layering and Stream Replication Video Multicast Schemes

Rate Variability

Page 39: A Comparison of Layering and Stream Replication Video Multicast Schemes

Quality Adaptation Framework

d

Ci[k]

Si[k]

time

cum

ulat

ive

data

in th

e ith

laye

r Xi[k]

k1 k2k0

select selectdiscard

C[k]: transmission resource constraintX[k]: cumulative data sizeS[k]: cumulative selected data sized: threshold

Page 40: A Comparison of Layering and Stream Replication Video Multicast Schemes

Optimal Quality Adaptation

Threshold should be equal to the receiver buffer size to achieve Minimum quality variability Necessary condition of maximum

bandwidth utilization

Page 41: A Comparison of Layering and Stream Replication Video Multicast Schemes

Online Adaptation

Estimating the threshold point without assuming the available bandwidth information in advance

The available bandwidth is estimated by an MA style linear estimator

Page 42: A Comparison of Layering and Stream Replication Video Multicast Schemes

Experiment Model

0

6

2

4

7

3

5

TFRC sender TFRC receiver

1

TCP sender TCP receiver

Page 43: A Comparison of Layering and Stream Replication Video Multicast Schemes

Bandwidth Variability

TCPTFRC

Page 44: A Comparison of Layering and Stream Replication Video Multicast Schemes

Performance over TFRC

Threshold-based streaming (Infocom’00)

Online adaptation

Page 45: A Comparison of Layering and Stream Replication Video Multicast Schemes

Performance over TCP

Threshold-based streaming

Online adaptation

Page 46: A Comparison of Layering and Stream Replication Video Multicast Schemes

Conclusion

Accommodated the mismatch between the rate variability and the bandwidth variability

Developed an optimal quality adaptation scheme for MPEG4 FGS video to reduce quality variation

Investigated the perceptual quality of different algorithms and options