IP Video Quality Testing Oct 2008 Agenda Video

  • Slides: 29
Download presentation
IP Video Quality Testing Oct 2008

IP Video Quality Testing Oct 2008

Agenda • Video Networks • IP Video Qo. S • Qo. S and Qo.

Agenda • Video Networks • IP Video Qo. S • Qo. S and Qo. E mapping vs Network Segments • IPTV testing • Trouble-shooting • Noise Analysis • Summary 2

Service Delivery ONT: Optical Network Terminal RG: Residential Gateway 3 ONU: Optical Network Unit

Service Delivery ONT: Optical Network Terminal RG: Residential Gateway 3 ONU: Optical Network Unit VOD: Video On Demand OLT: Optical Line Terminal IAD: Integrated Access Device

Triple Play Network/Service Challenges Video Hub Signaling and Core Ntwk Complex Network Environments require

Triple Play Network/Service Challenges Video Hub Signaling and Core Ntwk Complex Network Environments require Head Media Gateways to. End minimize expensive dispatches Ø Poor content Head End quality from content provider. ØMistakes here catastrophic Last Mile Cust Prem Service Assurance Systems ØSignal degradation. Ø Customer Data Traffic victimizes Vo. IP/Video Ø Mis-provisioning Hub Office ØImpulse noise creates Vo. D D/A ØCopper pair: incorrectly ØMisprovisioned imbalanced lines, Ø Misconfigured routed calls tributaries and ØBridge taps, etc. Premise Ø PCR Jitter from POTS routes Eth Aggr encoder, splicer Ø Splitter. Equipment IPSignaling MPLS/ ØData applications or VOD source. translations and ATM Core Network don’t experience RG Ø Firewalls incompatibilities. ØRouting/Link over STB TV Gb. E utilization…. poor these problems Ø PID mappings or Gb. E LOCAL ADS traffic engineering. because of TCP. Ø Inside wiring TVtable DVB data Ø Echo Canceller & CONTENT DEMUX x. DSL Cat 3, Coax RF mappings from configurations QT-600 PC ØPoor circuit ØVoice and video source QT-1100 routing…. . creates experiences Ø In-home Ø IGMP and RTSP QT-1100 QT-200 excessive delays issues because of Networking, Ø Proper Trans. Latency real-time nature. HPNA, Mo. CA, coding Wireless, BPL 15 -20% of Problems 4 Net. Complete 5 -10% of Problems 25% of Problems HST-3000 50% of Problems

blurring, edge distortion, visual noise Content Quality Qo. S and Qo. E mapping for

blurring, edge distortion, visual noise Content Quality Qo. S and Qo. E mapping for Video. Audio: Services Lip sync, drop Transport Quality outs Video Stream Quality Qo. E Parameters Pixelization, tiling frame freezes, Blue screen Transaction Quality Note 1: requires payload decode analysis PSI Table Data (Error) Service Provisioning PCR Jitter IGMP Latency Continuity Error RTSP Latency TCP Re-transmissions RTP Packet Jitter RTP Packet Loss IP Packet Jitter IP Packet Loss Error Indicator Count Video Qo. S parts 5 Service accessibility, Channel Change latency Pause, Play Latency

IP Video Network Segment Test Focus-Broadcast HE Qo. S Network Qo. S Access Qo.

IP Video Network Segment Test Focus-Broadcast HE Qo. S Network Qo. S Access Qo. S Content Quality Local Content Insertion Video Stream Quality Transport Quality Transaction Quality PCR Jitter, PSI Data, Error Ind, Compression – GOP, Type of Coder, Bit Rate, Profile Content Quality PCR Jitter Packet Loss Packet, Jitter Distance, Period 6 Total Qo. S @ STB IGMP Latency, PCR Jitter, PSI data, Error Ind, BW, MPEG-TS Pkt Loss, Jitter, Period, Distance, MOS

IP Video Network Segment Test Focus-VOD Hub Office Qo. S Content Quality Video Stream

IP Video Network Segment Test Focus-VOD Hub Office Qo. S Content Quality Video Stream Quality Network Qo. S Content Quality Video Stream Quality Transport Quality Transaction Quality PCR Jitter, PSI Data, Compression (GOP, Type of Coder, Bit Rate, Profiles), Trick Play Latency 7 Total Qo. S @ STB Access Qo. S Transaction Quality PCR Jitter Packet Loss Packet Jitter Distance, Period RTSP (Trick) Latency, PCR Jitter, PSI data, , BW, MPEG-TS Pkt Loss, Jitter, Period, Distance, MOS

IP TV Installation Test Set § STB/Modem emulation § IP Video Qo. S check

IP TV Installation Test Set § STB/Modem emulation § IP Video Qo. S check § Broadcast & VOD MDF STB Gathering DSLAM Network (IP) ATU-R Core Network (ATM, IP) TV Program Broadcaster Providers IGMP signal IP Local Content Added Video Service Platform Network Specific Thresholds 8

x. DSL Verification TE Replacement x. TU-C Replacement ISP ATM/IP 1 1 x. TU-R

x. DSL Verification TE Replacement x. TU-C Replacement ISP ATM/IP 1 1 x. TU-R Replacement DSLAM § Focus on showing end service functionality • Connectivity to the DSLAM • Connectivity to ATM network • Connectivity to IP network and beyond 9 RAS IP Routing

Video Qo. S – Network Specific Thresholds can be set in the HST for

Video Qo. S – Network Specific Thresholds can be set in the HST for Pass, Fail, Marginal The Qo. S screen shows the critical quality parameters for a video flow: PCR jitter measures this key parameter; if high the decoder can not properly decode the video payload: Pass < 100 m. S Latency is the IGMP latency, the time to change channels for Broadcast Video: time from IGMP request to Rx of first video packet: <200 m. S If Error Indicator count = >0 then content problem Cont. Err is the analysis of video Transport Stream packets which show the Continuity Error indicator set which = lost packet events: < 0. 1% If all items = Pass, no further analysis necessary by field tech. 10

MPEG over Ethernet Transmission issues 1) Perfect Video Stream Inter Frame delay OK typically

MPEG over Ethernet Transmission issues 1) Perfect Video Stream Inter Frame delay OK typically 2 -4 ms STB Buffer Video rate i. e 3 Mbps 2) Stream with jitter or rate under run Inter Frame delay to long MPEG Decoder Buffer underflows 3) Stream with jitter and/or rate over run Inter Frame delay to short Buffer overflows Drop Packet Ethernet packet contains up to 7 MPEG packets MPEG packet 188 Bytes 11

IP Video Testing Critical Diagnostic Fault Conditions: § Video impairments – – § Time

IP Video Testing Critical Diagnostic Fault Conditions: § Video impairments – – § Time of day specific – – § A pattern? Random throughout the day? Temporal component – – 12 All channels? One channel? Bursts? Randomly spaced?

IP Video Testing IP Fault Resolution: Step #1 § Video pixelizations – All channels?

IP Video Testing IP Fault Resolution: Step #1 § Video pixelizations – All channels? = yes • § Lost packets – Analyze physical layer stats • • • – Ethernet I/F stats ATM stats DSL stats Sectionalization • 13 Lost packet events will be seen on more than one stream If no errors are seen at the physical layer of both I/F ‘s, then packet loss is up-stream of the DSLAM

IP Video Testing IP Fault Resolution: Step #2 § Video pixelizations – – Physical

IP Video Testing IP Fault Resolution: Step #2 § Video pixelizations – – Physical layer problems? = yes Check DSL stats for errors • • If yes, impulse noise most likely cause, if noise margin at least 6 d. B. Review Bits/Tone graphs for notches – – Notches = Noise problems Check packet layer stats • • • If CRC errors = loop noise problems If Length errors, but no CRC then may be network Qo. S /Buffer over flow issues, not loop noise problems If both most likely loop noise issues Evaluate copper loop for narrow band wide band issues. Use temporal input to pinpoint sources. 14

IP Video Testing IP Fault Resolution: Step #2 § Video pixelizations – – Physical

IP Video Testing IP Fault Resolution: Step #2 § Video pixelizations – – Physical layer problems? = yes Check VDSL stats for errors • • • FEC errors: Un-correctable FEC errors If yes, impulse noise most likely cause, if noise margin at least 6 d. B. Review Bits/Tone graphs for notches – – Notches = Noise problems Check Packet stats – – – • RTP Stats RFC 3357 MDI If both most likely loop noise issues Evaluate copper loop for narrow band wide band issues. Use temporal input to pinpoint sources. 15

IP Video Testing IP Fault Resolution: Step #3 § Video pixelizations – § All

IP Video Testing IP Fault Resolution: Step #3 § Video pixelizations – § All channels? = no Lost packets – Analyze lost packet stats • § Typically none seen when trouble is one ch. Analyze PCR jitter stats • PCR jitter problems are typically source issues: – – § Typically a content issue Error Indicator Count – 16 Head End trans-coding problems Local Ad insertion problems No Analyze PSI Error stats – § Yes Analyze count • Corrupted content leaving the encoder

IP Video Testing IP Fault Resolution: Step #3 § Analyze PCR jitter stats •

IP Video Testing IP Fault Resolution: Step #3 § Analyze PCR jitter stats • PCR jitter problems are typically source issues: – – • § Report PCR Jitter stats with Time-of-Day, and Channel or Program ID PSI table data analysis – PAT and PMT data present • • 17 Head End trans-coding problems Local Ad insertion problems Present at or above a threshold rate, typically every half second, but in MSTV network usually only every second or 1. 5 seconds For each video program stream the PID’s (Packet Identifiers) are shown for each piece: Video, Audio, & PSI table data. The Type and Description data is decoded from the stream as received. Unknown data includes data not included in the other categories such a Fill/PAD data or data marked as “reserved”.

Impulse Noise Standards Definition § VDSL is more susceptible to impulse noise events due

Impulse Noise Standards Definition § VDSL is more susceptible to impulse noise events due to it’s use of a wider frequency spectrum than ADSL. Noise sources are being analyzed in several forms: – REIN (Repetitive Electrical Impulse Noise) • Less than 1 ms in duration • No bit errors desired • INP mitigation – PEIN (Prolonged Electrical Impulse Noise) • 1 to 10 ms in duration • No bit errors desired • INP mitigation – SHINE (Single Isolated Impulse Noise Event) • Duration greater than 10 ms • Due to duration, bit errors will typically occur • No loss of sync is desired 18

Impulse Noise Analysis Sources of Impulse Noise § § § § 19 Central heating/air

Impulse Noise Analysis Sources of Impulse Noise § § § § 19 Central heating/air - switch on/off A neighbour’s Jacuzzi (Motor) Television sets Impulse count with 3 d. B Video recorders below and above PC scanner threshold Faulty street lights. Proper Filter Faulty Metal Halide lighting Faulty fluorescent lighting External Hi/Lo Security Lights External low voltage lights Event capture Seasonal Lights Satellite Receiver AC Power faults near copper loops 00: 06 5 15 15

Spectral Noise Analysis Up to 5 MHz Max Hold shows transient peak values Up

Spectral Noise Analysis Up to 5 MHz Max Hold shows transient peak values Up to 15 MHz Up to 30 MHz Spectral noise analysis down to -145 d. Bm/Hz 20 Wider VDSL spectrum opens loop to new noise sources

HPNA/VDSL in Home – VDSL 2/Cu Noise/TDR HST as a troubleshooting tool § HST

HPNA/VDSL in Home – VDSL 2/Cu Noise/TDR HST as a troubleshooting tool § HST connects to TWP or Coax, HPNA or VDSL § Syncs to VDSL 2, monitors copper, TDR analysis § Monitors 0 -30 MHz spectral analysis: Transient and Impulse noise § Interactive or long term monitoring 0 -30 MHz Copper Noise CPE VDSL 2 Sync Analysis STB TWP SAI IP DSLAM B-Box / SAI 21 VDSL Splitter Coax Adapt HST-3000

IP Video Testing IP Fault Resolution: Step #4 MDF Gathering DSLAM Network (IP) STB

IP Video Testing IP Fault Resolution: Step #4 MDF Gathering DSLAM Network (IP) STB Core Network (ATM, IP) ATU-R § Excessive IGMP Latency – All channels? = no • – – Report as program source issue IP All channels? = yes Analyze time of day issues • • Random or a pattern Report issues – 22 IGMP signal Review: » Review IGMP snooping hierarchy » Review traffic loading, special event issues » Multi-cast flows, are all channels available at DSLAM » Temporal caching plan impacts Multi-cast Platform

Video Qo. S – Packet Loss Correlation IP Fault Resolution: Step #5 The Pkt

Video Qo. S – Packet Loss Correlation IP Fault Resolution: Step #5 The Pkt Loss screen shows the status of the physical layer: ATM and DSL layer measures provide insight into the ADSL loop performance: Errors here indicate packet loss may be due to loop problems. If no errors are seen on the access loop, then packet loss is North-bound of the DSLAM (in the network). If no errors, then problems are typically in the network. 23

Video Qo. S Bandwidth & Pkt. Analysis IP Fault Resolution: Step #6 Individual bit

Video Qo. S Bandwidth & Pkt. Analysis IP Fault Resolution: Step #6 Individual bit rates are shown for each portion of the video stream. Unknown data defined as any data, such as PAD bits to maintain a constant data rate, or private IP data over MPEG, that is detected is also shown. Packet stats parameters are measured at the IP layer. IGMP Latency is the measure of the network components to complete a program change, i. e. changing the channel in Broadcast video. 24

MOS Score Evaluation IP Fault Resolution: Step #7 MOS score analysis for video and

MOS Score Evaluation IP Fault Resolution: Step #7 MOS score analysis for video and audio portions: Video factor is like audio R-Factor showing a score built from network elements of the factors effecting the MOS score. The same analysis is provided for the audio portion of the program. 25

PID Map & Combined Stream Data IP Fault Resolution: Step #8 Totals for up

PID Map & Combined Stream Data IP Fault Resolution: Step #8 Totals for up to 3 streams: Combined total data rate for the streams under analysis, in this case 3. Simultaneous analysis of multiple streams is helpful in locating problem sources. 26

IPTV troubleshooting Putting Physical Analysis Together Distribution box ADSL Example: DSLAM Street cabinet MDF

IPTV troubleshooting Putting Physical Analysis Together Distribution box ADSL Example: DSLAM Street cabinet MDF Not twisted Not shielded pair NID 1330 m ADSL tests Passed Copper tests Passed 280 m 18 m Copper tests Failed (impulsive noise, spectral) ADSL tests Failed (ADSL errors within UP stream) Copper tests Failed (impulsive noise, spectral) 27 ADSL 2+ modem

IP Video Testing Focus Summary – Video Services (Broadcast & VOD ) Verification and

IP Video Testing Focus Summary – Video Services (Broadcast & VOD ) Verification and Quality analysis at the Video Packet Level and the Transport Stream (RTP/IP) level – Features: • Service verification: Broadcast and VOD services – GMP signaling emulation for Broadcast Video – RTSP signaling emulation for VOD • Video Transport Steam packet flow analysis – Video Qo. S, PID map, PSI analysis, – Packet Loss and Jitter and correlation with physical layer quality • Modes: – Terminate: Res Gateway or STB emulation – Monitor: Through or Ethernet Bridge – Integration into Process Improvement programs • Store and export results • Standard and custom scripts 28

. Testing the Physical Medium is Not Enough JDSU Application-Aware Solutions TM § Equip

. Testing the Physical Medium is Not Enough JDSU Application-Aware Solutions TM § Equip work force for IP service testing as well as circuit testing § Minimize customer site visits with centralized fault location § Meet customer due dates – get to revenue quickly § Ensure IP service Qo. S metrics are met producing good Qo. E – reduce churn Physical Layer Network Layer Data Link Layer Application Layer IP Vo. IP Physical Medium Video IP 29 Voice Content ftp HTTP IP Data Signaling VOD