Download presentation
Presentation is loading. Please wait.
Published byNaomi Moody Modified over 9 years ago
1
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 Efficient Path Aggregation and Error Control for Video Streaming OMESH TICKOO, Shiv Kalyanaraman, John Woods Rensselaer Polytechnic Institute (RPI) Sponsors: ARO, DARPA-NMS, Intel : “shiv rpi”
2
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 2 Introduction q Motivation: Video over best-effort Internet q Broadband => more access bandwidth q End-to-end (E2E) => constraints due to path congestion q Virtual extension of broadband access pipe E2E using multi-paths q Path Diversity: dimensions q Aggregate Capacity q Delay diversity q Loss diversity q Correlations in path performance characteristics q Key: Match inherent content diversity to path diversity
3
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 3 Motivation: Internet Path Congestion limits E2E bandwidth Internet Server Access Link Client Access Link Performance Access Link Speed Performance Saturation (even w/ many flows/path)
4
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 4 Multi-paths? Overlays or peers can provide path diversity even if multi-paths not available natively in the Internet. Issue: diversity of performance (b/w, delay, loss), possible correlations…
5
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 5 Path (Flow) Aggregator/ Multiplexer Path (Flow) Aggregator/ De- multiplexer Internet E2E Broadband Virtual Pipe Abstraction!! Server Access Link Client Access Link Performance Access Link Speed Smart Multi-path Capacity Aggregation (SMCA): Motivation Performance Scaling
6
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 6 Time Lossy Low Capacity High Delay/Jitter Network paths usually have: low e2e capacity, high latencies and high/variable loss rates. Single path issues: capacity, delay, loss…
7
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 7 SMCA: Leverage Diversity! Low Perceived Delay/Jitter Low Perceived Loss High Perceived Capacity
8
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 8 Delay Diversity Unit Loss Diversity Unit Network Receive Buffer Content SMCA: Framework
9
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 9 Paths Ranked by LatencyApplication Data Low Delay RANK High Delay RANK SMCA: Delay Diversity Unit
10
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 10 Paths Ranked by LatencyApplication Data Low Delay RANK High Delay RANK Early deadline packets mapped to low-delay paths SMCA: Delay Diversity Unit
11
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 11 Paths Ranked by LatencyTransmit Queue Low Delay RANK High Delay RANK Early deadline packets (in order of rank) mapped to low-delay paths (in order of rank) SMCA: Delay Diversity Unit
12
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 12 Paths Ranked by LatencyTransmit Queue Low Delay RANK High Delay RANK Late deadline packets mapped to high-delay paths… Note: these packets leave the sender roughly at the same time as the early-deadline packets SMCA: Delay Diversity Unit
13
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 13 Paths Ranked by LatencyTransmit Queue Low Delay High Delay Consider a delay-based group of paths and the associated packets… SMCA: Delay Diversity Loss Diversity
14
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 14 Paths Ranked by LatencyTransmit Queue Low Delay High Delay Consider a delay-based group of paths and the associated packets… SMCA: Delay Diversity Loss Diversity
15
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 15 Paths Ranked by Loss Rate n GOPs Low Loss RANK High Loss RANK Re-rank Paths within this group based upon packet loss rates SMCA: Loss Diversity Unit
16
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 16 I Paths Ranked by Loss Raten GOPs Low Loss RANK High Loss RANK Enlarged View of Packets (with content labels) and Paths P B B P B B SMCA: Loss Diversity Unit
17
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 17 I Paths Ranked by Loss Raten GOPs Low Loss RANK High Loss RANK P B B P B B Map high priority packets (eg: I-frame packets) to low loss rate rank paths SMCA: Loss Diversity Unit
18
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 18 I Paths Ranked by Loss Raten GOPs Low Loss RANK High Loss RANK P B B P B B Continue map packets to low loss rank paths based upon priority (Eg: P-frames get the next set of loss-ranked paths) SMCA: Loss Diversity Unit
19
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 19 I Paths Ranked by Loss Raten GOPs Low Loss RANK High Loss RANK P B B P B B Lowest priority packets get high loss rate paths (within the delay-based group of paths) SMCA: Loss Diversity Unit
20
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 20 I Paths Ranked by Loss Raten GOPs Low Loss RANK High Loss RANK P B B P B B FEC (unequal FEC) for a GOP mapped within the same delay-group, but mapped to the higher loss paths SMCA: Loss Diversity Unit + FEC I-FEC P-FEC
21
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 21 SMCA: Performance with increasing number of Paths Num. Of Paths 1 2 3 4 5 PSNR (dB) 20.98 22.48 25.42 26.02 28.04 Table 1. Average PSNR Variation with Number of Paths Background traffic generator Background traffic sink Content SourceContent Sink
22
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 22 Topology to test delay diversity and loss diversity gains Content SourceContent Sink Background traffic generator Background traffic sink 5 paths
23
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 23 SMCA gains with delay diversity Avg. Delay (ms) SMCA PSNR(dB) PT PSNR(dB) OPMS PSNR(dB) 30021.7818.7311.03 10025.1224.2119.19 5028.3229.4624.33 3030.1231.6327.96 Table 3. Gains with Delay Variation SMCA achieves even better performance (than simple multi- path mapping: OPMS) when average delay and jitter is higher
24
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 24 SMCA gains with loss diversity Avg. Loss Prob. SMCA PSNR(dB) PT PSNR(dB) OPMS PSNR(dB) 0.422.7820.3111.64 0.3526.3226.8618.21 0.129.0329.0224.43 0.0529.3231.8226.06 Table 2. Gains with Loss Variation SMCA achieves even better performance (than simple multi- path mapping: OPMS) when average loss and loss variations are higher!
25
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 25 Summary q Multi-path performance diversity can be leveraged E2E q Key: must be mapped to content diversity q (Similar to lessons learnt from content-driven unequal FEC protection vs uniform FEC protection) q Ideas: q Map late deadline packets to high latency paths q Map higher priority packets to lower loss rate paths (within a delay- based group of paths) q FEC packets sent on paths different from that of associated content (FEC: lower priority) q Our scheme can scale to handle lots of paths q Possible with p2p networks (eg: 10-100 kbps from single path, but 10s of paths) q Does not require MD coding, or high complexity optimization
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.