Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

13
Rajesh J. Vale Department of Computer Science 2009 IPTV Rajesh Vale

Transcript of Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Page 1: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

IPTV

Rajesh Vale

Page 2: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Objective

Use existing infrastructure/technology for– Access

– Transport

– Video encoding

– Authentication

– Billing

To provide Television services (Live & VoD) to consumers.

Page 3: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Access

Wireless feasible? xDSL

– 512kbps, 768kbps etc is inadequate Solution?

• Fiber to home• Fiber to neighborhood

Mobility Vs Nomadicity– Mobility: Point of attachment changes in-session

Page 4: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Transport

Efficient way of transporting in IP layer– Broadcast TV (bTV)

– VoD

– Live-Pause TV or Network PVR bTV natural candidate for Multicast VoD for Unicast Live-pause TV?

Page 5: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

bTV: Multicast Issues

What about lost packets?– Retransmission– FEC– Interpolation

Is there such a thing as ‘reliable multicast’?

Multicast tree administration?

Page 6: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Video Encoding

Group of Pictures (GOP)– GoP size varies over time

– Average with variance bounds exists I/P/B frames

– Transmission sequence : IBBPBBPBBPBB CBR Vs VBR

– CBR is preferred for predictability and efficient network engineering

Page 7: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

bTV: QoE

Channel surfer experience– Load on network

Immediate channel change– How to overcome wait for I-Frame?

• Smaller GoP encoding• Start from previous I-Frame

– Catchup?

• Create I-frame on request

How network knows I-frame carrying IP packet?

Page 8: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

VoD

Buy now, watch anytime in next 30 days Trick mode operations

– Pause/FF/Rew Content theft prevention

– DRM system on top of every content flow

Page 9: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Authentication

Identity levels– Household

– Individual

– Avatar How to use existing AAA infrastructure?

– Limitations 802.1X, SIM card, PKI certificate…

Page 10: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Billing

Online billing– Non-IPTV services

Offline billing– Traditional billing

Page 11: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Future IS/IT

IP address is associated with viewer. Leverage it for unique services

Websurfing/shopping Targeted Advertisements Interactive TV Blended services

– Caller ID on TV– E-mail alerts

WEBSERVICES

Page 12: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

MSTV Architecture

http://www1.alcatel-lucent.com/com/en/appcontent/opgss/IPTV_sb_tcm228-693051635.pdf

Page 13: Rajesh J. ValeDepartment of Computer Science 2009 IPTV Rajesh Vale.

Rajesh J. Vale Department of Computer Science 2009

Cisco IPTV Architecture

http://www.cisco.com/global/DK/docs/presentations/partnere/IPTV-Copenhagen-291105.pdf