the compassionate, the merciful
S. Kasaei S. Kasaei Room: CE 307 Department of Computer Engineering Sharif University of Technology y gy E ‐ Mail: skasaei@sharif.edu Webpage: http://sharif.edu/~skasaei Lab. Website: http://ipl.ce.sharif.edu p // p
Most of the slides used in this lecture have been provided by: Dr. Dapeng Wu (University of Florida, Department of Electrical &Computer Engineering) based on the book: Video Processing & Communications written by: Yao Wang, Jom Ostermann, & Ya ‐ Oin Zhang Prentice Hall 1 st edition 2001 ISBN: 0130175471 Prentice Hall, 1 edition, 2001, ISBN: 0130175471. [SUT Code: TK 5105 .2 .W36 2001].
Scalable Video Transport p over Wireless IP Networks
Outline Outline � Challenges for quality video transport Challenges for quality video transport � Architecture for Video Streaming Systems � Video Compression Video Compression � Application ‐ Layer QoS Control for Streaming Video � Continuous Media Distribution Services � Streaming Services � Media Synchronization � Protocols for Streaming Video � Streaming Video over Wireless IP Networks � Summary S Kasaei Kasaei 6
Lecture Outline Lecture Outline • Challenges for video over wireless IP Challenges for video over wireless IP networks. • An adaptive framework for video over wireless • An adaptive framework for video over wireless IP networks: – Scalable video representations. Scalable video representations – Network-aware end systems. – Adaptive services. Ad ti i • Summary Kasaei Kasaei 7
Bandwidth Fluctuations Bandwidth Fluctuations Access Access SW SW Domain B Domain A Domain A Domain C Access Access Internet Source SW SW Gateway Gateway Cellular Networks Wireless LAN 1 Mb/s Receiver 64 kb/s Mobile PC Mobile PC Mobile PC Kasaei Kasaei 8
Challenges Challenges • Unreliability: – Fading. – Noise. • Bandwidth fluctuations: – Moving between different networks (LAN to WAN). – Hand-off. – … • Heterogeneity for multicast. Kasaei Kasaei 9
Unicast vs Multicast Unicast vs. Multicast Unicast Multicast Kasaei Kasaei 10 10
Three Independent Techniques Three Independent Techniques • Scalable video coding Scalable video coding. • Network-aware adaptation of end systems: – Network awareness. N t k – Adaptation. • Adaptive QoS support from networks: • adaptive services. p Kasaei Kasaei 11 11
Scalable Video Representations Scalable Video Representations Layered video encoding/decoding. [D denotes the decoder.] Kasaei Kasaei 12 12
An Application: IP Multicast An Application: IP Multicast Kasaei Kasaei 13 13
Our Approach Our Approach • Unify the three techniques: Unify the three techniques: an adaptive framework Kasaei Kasaei 14 14
Network ‐ Aware End Systems Network ‐ Aware End Systems � Why using network ‐ aware end systems? y g y � All layers may get corrupted with equal probability without awareness of channel status. � How? � How? � Discard enhancement layers at the sender based on network status. � Network ‐ aware adaptation: � Network monitoring: collect information. � Adaptation: adapt video representations based on Ad t ti d t id t ti b d network status. Kasaei Kasaei 15 15
T Taxonomy of Network Monitoring f N k M i i Criteria Type of monitoring Method of monitoring Active Passive Monitoring frequency g q y On demand Continuous Replication of information Centralized Distributed Kasaei Kasaei 16 16
Ad Adaptation/Scaling t ti /S li An architecture for transporting scalable video from a mobile terminal An architecture for transporting scalable video from a mobile terminal to a wired terminal. Kasaei Kasaei 17 17
Scaling Scaling � The operations of a scaler: � Drop the enhancement layers. � Do not scale the video. � Scaling based on network status: � Available bandwidth. A il bl b d idth � Channel quality (BER). Kasaei Kasaei 18 18
Ad Adaptive Services ti S i � Objective: � Achieve smooth change of perceptual quality in presence of bandwidth fluctuations. f b d idth fl t ti � Functions: � Reserve a minimum bandwidth for the base layer � Reserve a minimum bandwidth for the base layer. � Adapt enhancement layers based on available bandwidth and the fairness policy. Kasaei Kasaei 19 19
Adaptive Services Adaptive Services � Provisioning: � Provisioning: � End ‐ to ‐ end deployment (our focus). � Local deployment. p y � Components: � Service contract. � Call admission control and resource reservation. � Substream scaling. � Substream scheduling. � Link ‐ layer error control. Kasaei Kasaei 20 20
Service Contract Service Contract � A service contract consists of multiple � A service contract consists of multiple subcontracts: � Bandwidth reservation for the base layer . y � No QoS guarantee for enhancement layers. � Enforcement: � Shaping. � Priority. Kasaei Kasaei 21 21
Call Admission Control (CAC) Call Admission Control (CAC) � Objective: Object e: � Provide a QoS guarantee while efficiently utilizing network resources. � The operation of CAC: check � Whether QoS for existing connections is violated. � Whether the incoming connection’s QoS can be met. Wh h h i i i ’ Q S b Kasaei Kasaei 22 22
Resource Reservation Resource Reservation � Different from the counterpart in wired networks: e e t o t e cou te pa t ed et o s: � The reserved bandwidth may not be rigidly guaranteed in wireless networks. � Two parts of resource reservation: � Reserve resources along the current path. � Reserve resource on the paths from the current base R h h f h b station to neighboring base stations. Kasaei Kasaei 23 23
Mobile Multicast Mechanism Mobile Multicast Mechanism • Objective: Objective: – Provide seamless QoS during a handoff. • Multicast mechanism: M lti t h i – Multicast the base layer to the neighboring base stations. t ti Kasaei Kasaei 24 24
Substream Scaling Substream Scaling � Objective: Object e: � Adapt video streams during bandwidth fluctuations and/or under poor channel conditions. � Scaling decision based on utility fairness or max ‐ min fairness: � Utility fairness is based on utility functions. Utilit f i i b d tilit f ti � Max ‐ min fairness is based on revenue. Kasaei Kasaei 25 25
Substream Scaling Substream Scaling An architecture for transporting scalable video from a wired terminal to a mobile terminal to a mobile terminal. Kasaei Kasaei 26 26
Utility Functions Utility Functions Kasaei Kasaei 27 27
Substream Scheduling Substream Scheduling Kasaei Kasaei 28 28
Link Layer Error Control Link ‐ Layer Error Control � Forward error correction (FEC). � Automatic repeat request (ARQ). � Truncated type ‐ II hybrid ARQ. � Delay constrained hybrid ARQ. � A receiver sends request based on delay bound of the packet. Kasaei Kasaei 29 29
Delay Constrained Hybrid ARQ Delay ‐ Constrained Hybrid ARQ RCPC : Rate compatible punctured convolution RCPC : Rate compatible punctured convolution. Kasaei Kasaei 30 30
Summary Summary � Objective: end ‐ to ‐ end solution to providing QoS for video transport over wireless IP networks. � Our approach: an adaptive framework � Scalable video representations. � Network ‐ aware end systems. � Adaptive services. Ad i i � Advantages of the adaptive framework: � Perceptual quality is changed gracefully. P t l lit i h d f ll � Resources are shared in a fair manner. Kasaei Kasaei 31 31
Homework 13 Homework 13 • Reading assignment: – Chap. 15 p • Computer assignment: – Course project Course project Kasaei Kasaei 32 32
Recommend
More recommend
