Transcript Temporal Error Concealment

On the Interactions Between Layered
Quality Adaptation and Congestion
Control for Streaming Video
11th International Packet Video Workshop
Nick Feamster
Deepak Bansal
Hari Balakrishnan
MIT Laboratory for Computer Science
http://nms.lcs.mit.edu/projects/videocm/
Not Like Watching TV!
April 30, 2001
MIT Laboratory for Computer Science
2
Why Is This Happening?
• The Internet poses several problems for the
delivery of data
– Variable Bandwidth
– Variable Delay
– Packet Loss
• Very detrimental to interactive video delivery
• How do we transmit video on the Internet
in the face of varying bandwidth?
April 30, 2001
MIT Laboratory for Computer Science
3
System Context
Request for Video
Streaming
StreamingVideo
Video
Loss/Latency Feedback
Video Server
Video Client
• This talk is about bandwidth adaptation
• Conclusion: The combination of smooth congestion control
and clever receiver buffering can overcome the evils of
bandwidth variation!
April 30, 2001
MIT Laboratory for Computer Science
4
Bandwidth Adaptation
• Available bandwidth varies with time
• Servers should adapt to varying
bandwidth
– Congestion Control: Transmission rate must
• correspond to available bandwidth
• be TCP-friendly
– Quality Adaptation: Quality of video should
correspond to transmission rate
• Limited capacity for buffering!
April 30, 2001
MIT Laboratory for Computer Science
5
Layered Video
• Simulcast: Each layer is independent
• Hierarchical: Higher depends on lower
– Base/Enhancement layers
– Linear granularity (C bits/layer)
Simulcast Layering
April 30, 2001
Hierarchical Layering
MIT Laboratory for Computer Science
6
Binomial Congestion Control
w(t)
AIMD
t
AIMD
Increase
Decrease
Binomial
a
bw
a / wK
b wL
• Trade-off between increase aggressiveness and decrease magnitude
• K+L=1 implies TCP-friendly [Bansal, INFOCOM 2001]
• SQRT has a modest backoff (~R1/2) => attractive for streaming media
April 30, 2001
MIT Laboratory for Computer Science
7
Reduced Oscillations
In many cases, AIMD drops
multiple layers in one backoff!
This is not the case with SQRT.
Rate oscillations in SQRT
are much less pronounced
than in AIMD.
SQRT
Bitrate
Bitrate
AIMD
April 30, 2001
MIT Laboratory for Computer Science
8
Layered Quality Adaptation
• Tailor video to available bandwidth!
• Can be immediate or receiver-buffered
– Rejaie et al., SIGCOMM ‘99
April 30, 2001
MIT Laboratory for Computer Science
9
Receiver Buffering
R
Consumption Rate
(na+1)C
C
Optimal
L0 buffering
R/2
• Allocate more buffer space to lower layers
• Add a layer when the following conditions are met:
– Enough bandwidth is available
– Enough video is buffered to sustain a backoff and continue
playing all of the layers (including the new layer)
April 30, 2001
MIT Laboratory for Computer Science
10
Interaction of SQRT and QA:
We Win!
R
Consumption Rate
(na+1)C
C
R–
bR1/2
Total buffering to add an additional
layer is O(R3/2) rather than O(R2)
• With SQRT:
– Smaller Oscillations
– Less buffering required for quality adaptation
April 30, 2001
MIT Laboratory for Computer Science
11
Reduced Buffering
56% less buffering to
add 4 layers
SQRT requires less
buffering to add layers!
April 30, 2001
MIT Laboratory for Computer Science
12
Conclusion
• Combination of SQRT congestion
control with receiver quality adaptation
enables smooth video delivery
– Reduces rate oscillations
– Reduces buffering/Increases interactivity
• Software is available
– Includes selective reliability for packet loss
– http://nms.lcs.mit.edu/software/videocm/
April 30, 2001
MIT Laboratory for Computer Science
13
Extra Slides
April 30, 2001
MIT Laboratory for Computer Science
14
Outline
• Problem Overview
• Background
– Bandwidth Variation
– Quality Adaptation
– Binomial Congestion Control
• Approach
• Results
• Conclusion
April 30, 2001
MIT Laboratory for Computer Science
15
The Goal
• TCP-friendly congestion control
• Reduce rate oscillations:
– Limit size of playout buffer
– Smooth perceptual quality
• Limit receiver buffering for QA
– Reach acceptable playout rate faster
– More interactivity in certain cases (i.e., if RTT and
RTT jitter are small)
April 30, 2001
MIT Laboratory for Computer Science
16
Results of SQRT
• Tested on emulated network conditions with
Dummynet and SURGE toolkit
• SQRT reduces rate oscillations for:
– Immediate adaptation
– Receiver-buffered QA
• Also reduces buffering:
–
–
–
–
Less jitter due to rate oscillations
Backoffs less severe => less QA buffering
Can play out at higher layers more quickly
More interactivity
April 30, 2001
MIT Laboratory for Computer Science
17
RTSP
MPEG Server
loss/RTT
callbacks
data
MPEG Client
loss/RTT/requests
data
loss/RTT/requests
Internet
SR-RTP
CM
April 30, 2001
RTP/RTCP
RTP/RTCP
MIT Laboratory for Computer Science
18
System Architecture
RTSP
MPEG Server
data
MPEG Client
loss/RTT/requests
Layering callbacks
RTP/RTCP
loss/RTT
Internet
SR-RTP
data
loss/RTT/requests
RTP/RTCP
callbacks
CM
April 30, 2001
MIT Laboratory for Computer Science
19
Reduced Oscillations
In many cases, AIMD drops
multiple layers in one backoff!
This is not the case with SQRT.
Rate oscillations in SQRT
are much less pronounced
than in AIMD.
Layers Dropped
SQRT
Bitrate
Bitrate
AIMD
April 30, 2001
MIT Laboratory for Computer Science
20