- Yeungnam Univ. Adavanced Networking Technology
Download
Report
Transcript - Yeungnam Univ. Adavanced Networking Technology
2008 YU-ANTL Lab Seminar
QoS Support in a Wireless Home Network
(UPnP and WLAN Architecture overview)
December 23, 2008
Shahnaza Tursunova
Advanced Networking Technology Lab. (YU-ANTL)
Dept. of Information & Comm. Eng, Graduate School,
Yeungnam University, KOREA
(Tel : +82-53-810-2497; Fax : +82-53-810-4742
http://antl.yu.ac.kr/; E-mail : [email protected])
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
1
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Outline
Introduction
Related work
WLAN QoS
IEEE 802.11e EDCA
IEEE 802.11e HCCA
WMM
QoS management architecture for 802.11-based wireless networks
UPnP QoS Architecture
QoS-Enabled Home Network
Admission Control mechanism
Run-time monitoring, feedback, and adaptation
Demo scenario and results
Conclusion
References
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
2
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Introduction
Motivation
IEEE 802.11 Wireless LAN (WLAN) technologies are widely used in
homes to distribute the broadband Internet connection
Do not require installation of new cabling
Equipment required is relatively inexpensive
Capacity is enough to support traditional applications
– Web surfing, file downloading
Challenges
Converged home network environments
Data network is converging with the multimedia and entertainment
applications (voice, TV, gaming)
Require the network top support timely delivery of data
– Delay sensitive
Currently used WLAN technologies do not support QoS provisioning
Bandwidth limitation, radio channels are error prone (multipath,
shadowing, interference, and so forth)
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
3
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Introduction (2)
In this presentation
Review the state of the art technologies that can be used to provide
QoS in the converged home networks
WLAN QoS and UPnP QoS architectures
Proposed solution for providing QoS in home environment
Demo scenario and experimentation
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
4
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Related work
QoS mechanisms defined in IEEE 802.11e
Enhanced Distributed Channel Access (EDCA)
Based on IEEE 802.11 Distributed Coordination Function (DCF)
– Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA)
Backoff time depends on the access category (AC)
– BK (Background), BE (Best Effort), VI (Video) and VO (Voice)
Contention based access mechanism and cannot guarantee QoS in case of excess
traffic in the network
Hybrid Coordination Function (HCF) Controlled Channel Access (HCCA)
Based on Point Coordination Function (PCF)
Polled, contention-free access to the medium
– Hybrid coordinator (HC) takes control of the channel and assigns the transmission opportunities
to stations
Is not widely supported on the market due to its complex nature
Wi-Fi Multimedia (WMM)
Specification for WLAN QoS from the Wi-Fi Alliance
Uses EDCA as its access method
Most widely supported QoS mechanism on the market
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
5
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Related work (2)
Survey the main QoS enhancement schemes that have been proposed for
802.11 WLAN in [1]
Reviewed lack a management layer in the wireless network to react dynamically to the
continuous fluctuation of radio resource
QoS support for a mobile ad hoc network is presented in [2]
Solution requires some modifications to the link layer
[3] gives comparison of several existing QoS-aware middleware systems in
terms of
QoS specifications
QoS translation and compilation
QoS enforcement
QoS adaptation
Integrated middleware QoS framework is described in [4]
Requires the introduction of specific modules to both the end-hosts and the
intermediate nodes
Transparent QoS provisioning solution for AV services utilizing UPnP and
basic IEEE 802.11 MAC is presented in [5]
Media distribution system supporting QoS using UPnP QoS and IEEE
802.11p/Q priorities is presented in [6]
Some methods that monitor network conditions based on periodically
exchanging of status information and metrics among the nodes
Unnecessary resource consumption, and periodic unavailability of the shared
communication
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
6
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Related work (3)
UPnP QoS Architecture
Easy and seamless connectivity between standalone networked devices
from different vendors
Provides services for policy management and network resource
allocation
Middleware layer between the application and the network technologies
Does not define the actual implementation or methods for policy
enforcement, decision making, or QoS configuration
Three separate services that together comprise UPnP QoS framework
QoS Device
– Responsible of configuring the lower level QoS for the device it is associated
QoS Policy Holder
– Stores policies for the treatment of particular traffic types
QoS Manager
– Manages the QoS configuration of the network
– Receives QoS requests from control points, QoS Policy Holder
– Instructs the QoS Devices how to configure the network
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
7
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Related work (4)
UPnP QoS architecture
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
8
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Related work (5)
Games@Large (GaL)
European Information Society Technologies (IST) Integrated Project (IP)
intending to research, develop, and implement a novel platform for
gaming
Challenge game consoles by making digital games ubiquitously
available to users in environments
Platform is not restricted to be used by dedicated devices
GaL architecture will be based on distributing the game processing
between the game server and client devices
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
9
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
QoS-Enabled Home Network
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
10
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
QoS-Enabled Home Network (2)
Interaction diagram depicting interactions of agents involved in the
QoS management system
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
11
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
QoS-Enabled Home Network (3)
Admission Control mechanism
Performed by QoSManager
Decide if a new stream can enter the network and how many resources can be
allocated to it
Proposed procedure is based on the estimation of the end-to-end
available WLAN bandwidth
Estimate the available bandwidth by means of an active probing tool [7]
Common way to ado Admission Control
Reject requests by new streams to enter the network when the resources
are close to be depleted
Main idea of proposed Admission Control
Allow the new high priority stream to enter the network
– If there are some already running streams at low priority: an appropriate amount
of network resources can be revoked to low priority stream and reallocated to the
new high priority one
In order to revoke the usage of a portion of the allocated bandwidth fro a
stream, originating streaming server should be able to apply source
transrating
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
12
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
QoS-Enabled Home Network (4)
Admission Control mechanism (cont.)
QoSManager estimated the overall available bandwidth for reallocation
the available bandwidth - retrieved through the AST_GetAvailableBW
the actual aggregate throughput flowing through all the active QoSDevices
– QoSManager needs to know the amount of bandwidth which is really used by each
active QoSDevice
– Accomplished by invoking AST_GetQoSMetrics action
QoSManager has to distribute the bandwidth budget among the existing
streams and the new incoming one
Based their relative priorities
– Streams with higher priority could be allowed to run at full rate
– Streams with lower priority could be transtrated or even stopped
Final result of the Admission Control procedure
Share of already allocated resources could be released and reallocated to the
new incoming stream
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
13
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
QoS-Enabled Home Network (5)
Run-time monitoring, feedback, and adaptation
Run-time monitoring
To face the worsening of the wireless link quality while the A/V streams run
Even if AC ensures that new streams will now cause the saturation of the wireless
resources, it is still possible that new interfering sources arise after A/V stream
admission
Proposed solution is based on the notification of the values of relevant
metrics
If a QoSDevice received notices that the some QoS metrics falls below an alarm
threshold value, these events are propagated by means of UPnP QoS messages
to the QoSManager
– Indicate decay of wireless resources
– QoSManager could decide to collect other metrics by invoking the AST_GetQoSMetrics
» Try to compute a new set of streaming parameters
– New parameter set is propagated by calling the AST_SetTrafficQoSParameters
» Allowed but rate of a stream
When better network condition is available, the QOSManager is able to detect
improvements
– Periodically checking the channel status
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
14
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results
Scenario #1
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
15
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (2)
Admission Control mechanism
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
16
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (3)
Admission Control Mechanism (cont.)
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
17
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (4)
QoS continuous monitoring (QCM)
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
18
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (5)
Scenario #2
Laptops were equipped with D-Link DWL-G650 IEEE 802.11g WLAN
radios and MadWiFi Linux driver was used
Parameters
Station 1
– 5Mbps CBR UDP traffic (voice)
– From 0 s to 30 s
Station 2
– 20.5Mbps TCP traffic (FTP, background)
– From 10 s to 20 s
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
19
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (6)
Throughput results
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
20
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Demo scenario and results (7)
RTT
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
21
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Conclusion
In this presentation
Reviewed the state-of-the are technologies related to wireless networks
with QoS support for home networks
A framework to manage QoS for A/V streaming services in a WLANbased digital home networks has been presented
Optimally accomplish QoS requirements on the based of relevant metrics at
network and application level
Estimate the available WLAN bandwidth at any time
Propagate the relevant metrics
– Event notifications among the appropriate network devices
Capability of multimedia servers to change the video encoding rate on-the-fly
Not required to run on specific hardware/MAC enhancement
Experimental to confirm the adequacy of WMM to G2L architecture
Competing traffic did not affect the real-time application performance at all
However, WMM cannot support QoS in existence of excess load on the
highest priority
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
22
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
References
[1] Q. Ni, L. Romdhani, T. Turletti, “A Survey of QoS Enhancements for IEEE 802.11 Wireless LAN,”
Journal of Wireless Communications and Mobile Computing, Wiley. August 2004, Volume 4, Issue
5: pp. 547-566.
[2] L. Chen, W. Heinzelman, “Network Architecture to Support QoS in Mobile Ad Hoc Networks,” IEEE
International Conference on Multimedia and Expo, June 2004, Volume 3, pp 1715-1718.
[3] K. Nahrstedt, D. Xu, D. Wichakakul, B. Li, “QoS-aware Middleware for Ubiquitous and
Heterogeneous Environments,” IEEE Communications Magazine November 2001, Volume 39,
Issue 11: pp. 140-148.
[4] C. S. Ong, Y. Xue, K. Nahrsdedt, “A Middleware for Service Adaptation in Differentiated 802.11
Wireless Networks,” in Proceedings of the Workshop on Coordinated Quality of Service in
Distributed Systems, (COQODS) 2004.
[5] Y. S. Sun, C-C. Y, and M. C. Chen, “Content-aware automatic QoS provisioning for UPnP AV-based
multimedia services over wireless LANs”, Fifth International Conference on Computational
Science (ICCS), May 2005, Atlanta, GA, USA, pp. 444-452.
[6] S-G. Choi, D-O. Kang, J-W. Lee, “A media distribution system supporting priority-based QoS in a
home network”, The 8th International Conference on Advanced Communication Technology
(ICACT 2006), Feb 2006, Phoenix Park, Korea, vol 3, pp. 1532-1536.
[7] M. Jain and C. Dovrolis, “Pathload: A Measurement Tool for End-To-End Available Bandwidth,” in
Proceedings of Passive and Active Measurements (PAM) Workshop, March 2002, pp.14-25.
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
23
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008
Thank you for your attention!
Advanced Networking Tech. Lab.
Yeungnam University (YU-ANTL)
24
YU-ANTL Lab. Seminar
Shahnaza Tursunova, December 23, 2008