Your SlideShare is downloading. ×
IPTV (and Digital Cable TV) Performance Management
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Saving this for later?

Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime - even offline.

Text the download link to your phone

Standard text messaging rates apply

IPTV (and Digital Cable TV) Performance Management

849
views

Published on


0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
849
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
28
Comments
0
Likes
2
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. IPTV (and Digital Cable TV) Performance Management Alan Clark Telchemy Incorporated IIT VoIP Conference 2008
  • 2. Outline •  IPTV/ Digital Cable service architectures •  What do service providers need to know? •  The myriad aspects of IPTV performance •  IPTV performance monitoring architecture
  • 3. IPTV/ Digital Cable service Mix of scrambled/ unscrambled streams Customer Premise Scrambled video streams Video sources DRM 3rd party Digital video content Unscrambled video streams IP Set Top
  • 4. Video Frame Structure I B B P B Typical assumptions: - GoP length of 15 or 30 (have seen 250, can be variable) - I frame bandwidth ~ 40% of total bandwidth - P/B frame bandwidth ~ 3-5% of total bandwidth per frame Impact of high motion? - VBR - leads to bandwidth spikes - CBR - can lead to motion related blockiness
  • 5. Video Frame Structure and Packet Loss I B B P B Error in Error in I frame will P frame will propagate propagate through following through following P and B frames P and B frames Error in B frame will only impact this frame
  • 6. The myriad aspects of IPTV performance •  Transmission performance –  Equipment faults, limited bandwidth, unexpected traffic patterns, subscriber access or LAN problems •  Configuration or equipment errors –  Mismatch between encoder settings and decoder capabilities –  Equipment fault or unwanted behavior •  Non-optimal encoder configuration –  GoP too long or short –  Bandwidth consumption too high or variable –  Unexpected content •  Video loss - blank, frozen, noisy video –  Loss of video feed –  Excessive error rates on video feed –  Software bug in video encoder
  • 7. What do service providers need to know? •  Head-end perspective –  Are all streams conforming to specification? –  Do streams have active content? –  How is bandwidth being used? –  Impact of bandwidth limitation * content on quality? •  Subscriber perspective –  Network problems (loss, bandwidth…) affecting individual/multiple subscribers? –  Impact of codec configuration * transmission problems on quality? –  Overall user view of service provider? –  Overall user view of individual channels?
  • 8. IPTV performance monitoring Embedded agent Management Middleware in STB Video sources IPTV probe DRM 3rd party Digital Embedded probe video in router content
  • 9. IP Set Top/ Agent Architecture EPG & Graphics HDMI Menu Playout Comp Buffer DRM Video/Audio Descrambler Decoder S- Video UDP/RTP/MP2TS UDP/MP2TS UDP/RTP MP2TS SQlive Set Top Box Network Interface
  • 10. IP Set Top Agent Architecture Content related alerts Content Analysis Level of motion/detail Playout Buffer Compute User RTP, MPEG Compute Metrics Experience Emulator Compute PerFrame Frame Decode & PerFrame Perceptual Aggregation Per Quality Frame Quality Model & Quality Identification Reporting Error Propagation through GoP Time view FEC, R-UDP Model Channel view SQlive agent architecture Stream being watched by subscriber
  • 11. Head-end Probe Architecture Content related alerts Content Analysis Level of motion/detail Playout Buffer Compute RTP, MPEG Compute Metrics Emulator Compute PerFrame Frame Decode & PerFrame Perceptual Aggregation Per Quality Frame Quality Model & Quality Identification Reporting Error Propagation through GoP Time view FEC, R-UDP Model Channel view SQlive agent architecture All video streams
  • 12. Perceptual Quality Model Perceptual Quality Model Per-Frame Quality Estimates Video Qf(i), Qf(i+1)…… Integration Model I/B/P frame Bandwidth MOS-V Bandwidth Limitation MOS-A Model Perceptual Quality Content Estimate Content data Sensitivity Models Audio Audio quality Integration Estimates Model Af(i), Af(i+1)……
  • 13. Performance verification •  Main focus on subjective testing –  Expert panels of 16-20 viewers –  SD (480i) and HD (720p, 1080i and 1080p) tests –  Mix of DCR and ACR tests
  • 14. Scrambled vs Unscrambled Video •  Unscrambled •  Scrambled –  MOS-V = 3.1 –  MOS-V = 3.0 –  EPSNR = 27.5 –  EPSNR = 27.0 –  I frame packets –  I frame packets •  Received 6225 •  Received 6367 •  Lost 42 •  Lost 42 –  P/B frame packets –  P/B frame packets •  Received 20779 •  Received 20637 •  Lost 40 •  Lost 21 –  IP Statistics –  IP Statistics •  Loss rate = 0.23% •  Loss rate = 0.23%
  • 15. Summary •  IPTV/ Digital Cable service architectures •  What do service providers need to know? •  The myriad aspects of IPTV performance •  IPTV performance monitoring architecture