Video-on-demand broadcasting protocols. Jukka Leveelahti Tik Multimedia Communications

Similar documents
16.5 Media-on-Demand (MOD)

A variable bandwidth broadcasting protocol for video-on-demand

Combining Pay-Per-View and Video-on-Demand Services

An Interactive Broadcasting Protocol for Video-on-Demand

A Dynamic Heuristic Broadcasting Protocol for Video-on-Demand

Efficient Broadcasting Protocols for Video on Demand

A Proactive Implementation of Interactive Video-on-Demand

1. Introduction. SPIE/ACM MMCN2003, Santa Clara, CA, Jan An Efficient VOD Broadcasting Scheme with User Bandwidth Limit

Tabbycat: an Inexpensive Scalable Server for Video-on-Demand

An optimal broadcasting protocol for mobile video-on-demand

Improving Bandwidth Efficiency on Video-on-Demand Servers y

An Efficient Implementation of Interactive Video-on-Demand

Improving Video-on-Demand Server Efficiency Through Stream Tapping

Improving Server Broadcast Efficiency through Better Utilization of Client Receiving Bandwidth

Lossless VBR Video Broadcasting with User Bandwidth Limit using Uniform Channels

Trace Adaptive Fragmentation for Periodic Broadcast of VBR Video

A Video Broadcasting System

Seamless Workload Adaptive Broadcast

A Lossless VOD Broadcasting Scheme for VBR Videos Using Available Channel Bandwidths

SWITCHED BROADCAST CABLE ARCHITECTURE USING SWITCHED NARROWCAST NETWORK TO CARRY BROADCAST SERVICES

Multimedia Time Warping System. Akiko Campbell Presentation-2 Summer/2004

PRACTICAL LOSSLESS BROADCASTING SCHEMES FOR VARIABLE BIT RATE VIDEOS IN VIDEO-ON- DEMAND SERVICE

Pattern Smoothing for Compressed Video Transmission

Providing VCR Functionality in Staggered Video Broadcasting

Internet Protocol Television

IP Video driving more Users & Uses

White Paper. Video-over-IP: Network Performance Analysis

Analysis of Retrieval of Multimedia Data Stored on Magnetic Tape

FullMAX Air Inetrface Parameters for Upper 700 MHz A Block v1.0

1022 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 19, NO. 4, APRIL 2010

THE HIGH-BANDWIDTH requirements and long-lived

Network. Decoder. Display

Connected Broadcasting

Video-on-Demand. Nick Caggiano Walter Phillips

Video-on-Demand Broadcasting Protocols: A Comprehensive Study

ECEN689: Special Topics in High-Speed Links Circuits and Systems Spring 2011

Deploying IP video over DOCSIS

Chapter 10 Basic Video Compression Techniques

Skyscraper Broadcasting: A New Broadcasting Scheme for Metropolitan Video-on-Demand Systems

Delivering on demand Video services in cable environment over the DVB-C path

An Introduction to PHP. Slide 1 of :31:37 PM]

Bridging the Gap Between CBR and VBR for H264 Standard

Alcatel-Lucent 5910 Video Services Appliance. Assured and Optimized IPTV Delivery

Using the VideoEdge IP Encoder with Intellex IP

SMART TV SEEKS DUMB NETWORK FOR MARRIAGE

Implementation of MPEG-2 Trick Modes

Lehrstuhl für Informatik 4 Kommunikation und verteilte Systeme

THE CAPABILITY of real-time transmission of video over

Multimedia Communications. Video compression

EE141-Fall 2010 Digital Integrated Circuits. Announcements. Homework #8 due next Tuesday. Project Phase 3 plan due this Sat.

A COMPARATIVE ANALYSIS OF TAPE TECHNOLOGIES FOR MID-RANGE SYSTEMS AND SERVER APPLICATIONS

Content storage architectures

Implementation of an MPEG Codec on the Tilera TM 64 Processor

White Paper Customized IPTV Setups with TVCaster Server Appliances

Reduction of Clock Power in Sequential Circuits Using Multi-Bit Flip-Flops

Video coding standards

A Unified Approach for Repairing Packet Loss and Accelerating Channel Changes in Multicast IPTV

Robust Transmission of H.264/AVC Video using 64-QAM and unequal error protection

Multimedia Communications. Image and Video compression

VVD: VCR operations for Video on Demand

ATSC TELEVISION IN TRANSITION. Sep 20, Harmonic Inc. All rights reserved worldwide.

2 MHz Lock-In Amplifier

EN2911X: Reconfigurable Computing Topic 01: Programmable Logic. Prof. Sherief Reda School of Engineering, Brown University Fall 2014

Cost Analysis of Serpentine Tape Data Placement Techniques in Support of Continuous Media Display

ASNT_PRBS20B_1 18Gbps PRBS7/15 Generator Featuring Jitter Insertion, Selectable Sync, and Output Amplitude Control

EE141-Fall 2010 Digital Integrated Circuits. Announcements. Synchronous Timing. Latch Parameters. Class Material. Homework #8 due next Tuesday

Enabling home networking for digital entertainment TM. IEEE Presentation. March 2005

data and is used in digital networks and storage devices. CRC s are easy to implement in binary

Joint Optimization of Source-Channel Video Coding Using the H.264/AVC encoder and FEC Codes. Digital Signal and Image Processing Lab

COMP 249 Advanced Distributed Systems Multimedia Networking. Video Compression Standards

EL302 DIGITAL INTEGRATED CIRCUITS LAB #3 CMOS EDGE TRIGGERED D FLIP-FLOP. Due İLKER KALYONCU, 10043

Minimax Disappointment Video Broadcasting

Basic rules for the design of RF Controls in High Intensity Proton Linacs. Particularities of proton linacs wrt electron linacs

IP TV Bandwidth Demand: Multicast and Channel Surfing


Deploying IP video over DOCSIS

1. INTRODUCTION. Index Terms Video Transcoding, Video Streaming, Frame skipping, Interpolation frame, Decoder, Encoder.

Boundless Security Systems, Inc.

SWITCHED INFINITY: SUPPORTING AN INFINITE HD LINEUP WITH SDV

Interframe Bus Encoding Technique for Low Power Video Compression

EEC 116 Fall 2011 Lab #5: Pipelined 32b Adder

Performance Evaluation of Error Resilience Techniques in H.264/AVC Standard

BUSES IN COMPUTER ARCHITECTURE

Mobile TV broadcasting in Japan

A Video Frame Dropping Mechanism based on Audio Perception

17 October About H.265/HEVC. Things you should know about the new encoding.

Robert Alexandru Dobre, Cristian Negrescu

March 24, California Institute of Technology. US Government support acknowledged.

DEDICATED TO EMBEDDED SOLUTIONS

Interlace and De-interlace Application on Video

Understanding Compression Technologies for HD and Megapixel Surveillance

Using deltas to speed up SquashFS ebuild repository updates

Video Disk Recorder DSR-DR1000

Joint use of LTP and Erasure FEC for space environments (ECLSA 2.0)

Introduction. Packet Loss Recovery for Streaming Video. Introduction (2) Outline. Problem Description. Model (Outline)

MULTIMEDIA TECHNOLOGIES

Analysis of MPEG-2 Video Streams

HEVC H.265 TV ANALYSER

HELICAL SCAN TECHNOLOGY: ADVANCEMENT BY DESIGN

Internet Protocol Television

Transcription:

Video-on-demand broadcasting protocols Jukka Leveelahti 17.4.2002 Tik-111.590 Multimedia Communications

Motivation Watch any movie at home when ever you like MPEG-2 at least 4 MB per second Too expensive! Two ways to reduce costs Proactive: broadcasting Reactive: many approaches

Terms segment chunk of video, n of these in right order make entire video consumption rate Processing rate of video in STB b, unit of measure for VOD server bandwidth slot Time for STB to consume a constant-sized segment channel Each stream in VOD server Does not need to be of bandwidth b Each video can be distributed over several channels

Client requirements when channel bandwidth > b or STB listens to multiple channels => We need local storage Size of storage and number of channels are the two things to minimize with clever broadcasting protocols

Staggered broadcasting protocols Starting times for video are staggered evenly across certain n of channels video starts at every D/n (D=duration) mins = phase offset Not efficient for server, to cut phase offset double means doubling bandwidth Minimal requirements for client Can handle interactive VOD Example Canal Digital KIOSK

Pyramid broadcasting protocols Viswanathan, Imilienski (1995) each video is n segments, S 1,..,S n available bandwidth divided evenly to n channels C 1,..,C n i th segment of each video broadcasted on channel C i Size of segments grow geometrically using parameter à 1

Pyramid broadcasting protocols client waits for S 1 on channel C 1 and starts consuming to receive all the time, receiving S i must start before S i-1 finishes client will never experience a break when α = b /m, where b is bandwidth of each channel typical α is 2,5 channel 1 channel 2 channel 3 S 1 S 1 S 1 S 2 S 1 S 2 S 1 S 3 S 2 S 3 S 1 S 2 S 1 S 1 S 3

Pyramid broadcasting - performance more efficient than Staggered Broadcasting client waiting time decreases exponentially with bandwidth 2h video with 10b bandwidth per video => 12 mins vs. 2 mins client requirements are high clients have to listen > 1 channels at once bandwidth per channel is very high requires large storage size optimized versions followed..

Permutation-based Pyramid Protocol each channel divided into p à 1 subchannels for each video starts of segments evenly staggered on subchannels client listens only one subchannel at a time need of storage down to third comparing to basic Pyramid Broadcasting cost: more bandwidth for same waiting times

Skyscraper Broadcasting Protocol replaces geometric series for determining amount of data on each channel each video divided into n equally sized segments number of consecutive segments to place on each channel determined by series { 1,2,2,5,5,12,12,25,25,52,52, } Equals of about 1,5 Each channel requires only bandwidth b, can use much more channels Width of channel is constrained, no need of storage to store the last (large) block in last channel

Skycraper Broadcasting Protocol 1998 improvements: dynamical scheduling of channels and more efficient segment-tochannel series in total, low transfer rates and storage needs while reducing also waiting times (found in Pyramid Broadcasting) low transfer rate at client causes waste of bandwidth in server

Fast Broadcasting Protocol (1997) opposite approach to Skyscraper Broadcasting Series is {1,2,4,6,8,16,32,64,..} Very low waiting times Clients receive all data from all channels at once, leads to high transfer rate and high need of storage (up to half of video length)

Pagoda Broadcasting Protocol (1999) goal to broadcast segments infrequently while maintaining even transfer rate to client uses series like predecessors {1,3,5,15,25,75,125,..} Big difference: segments don t need to be consecutive on channels Uses pairs of channels when assigning segments

Pagoda Broadcasting Protocol client waits for instance of S 1 on channel C 1 while consuming S 1 starts receiving from every other channel dedicated to that video each segment S i is broadcasted at least once every i slots of time client will have the segment ahead in buffer or receive directly from server when needed Still requires storage for about half of video Notice: pyramid protocols don t work with interactive VOD

Harmonic broadcasting protocols first Juhn & Tseng (1997) Each video divided into n equally sized segments S 1,..,S i These are continuously broadcasted in their own channels S i is broadcasted in channel C i with bandwidth b/i Sum of channel bandwidths is n i= 1 b i n = b i= 1 b i = bh ( n) H(n) is the harmonic number of n, hence the name

Harmonic broadcasting protocols series grows very slowly can use hundreds of segments without not much bandwidth example: with 5b 1,5 mins for 2h video local storage is needed about 37% of the video contains a bug fixed with Delayed Harmonic Broadcasting Protocol with twice the waiting time..

Harmonic broadcasting protocols 1998 three variations Cautious Harmonic Broadcasting Protocol C 1 not changed, C 2 alternates S 2 and S 3 C i from 3 to n, broadcasts S i+1 at bandwidth b/i b/2 more bandwidth than Delayed Harmonic Protocol but waiting time only 1 slot Quasi-harmonic Broadcasting Protocol segments are divided into fragments which are not broadcast in order waiting time still 1 slot, bandwidth converges to bh(n) as n of subsegments increases

Harmonic broadcasting protocols Polyharmonic Broadcasting forces client to wait m slots before consuming clients can receive while waiting, segments can be brodcasted with lower bandwidth (compared to Harmonic Broadcasting) can use m times as many segments, waiting time does not increase uses less bandwidth for a given waiting time than Quasiharmonic Broadcasting no interactive VOD with Harmonic Broadcasting

Summary - VOD server different protocols share the same strategy: if some videos are more popular than others and clients have local storage, then later parts of video can be broadcasted not as often as the earlier parts protocols can save bandwidth on VOD server and allow more videos or allow server to be cheaper

Summary - client requirements Broadcasting Protocol Storage requirement Bandwidth requirement Staggered (% of video) (multiples of b) 0 1 Pyramid Permutation-based Skyscraper Fast Pagoda Harmonic 75 20 10 50 45 40 4-5 alpha 2-3 2 6-8 5-7 5-6

Summary Comparing server and client requirements ther is no clear winner For example: Polyharmonic has lowest bandwidth requirements on server, but too many data channels per video Pagoda is easy on server also, but client bandwidth too high Staggered Broadcasting still only one for interactive VOD and no extra load on client

Open questions and research interactive VOD protocols assume fixed bit rate, not the case with MPEG changes in video popularity are difficult to handle Staggered model still the easiest

Questions? Thanks!