City University of Hong Kong
DSpace
 

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Computer Science  >
CS - Master of Philosophy  >

Please use this identifier to cite or link to this item: http://hdl.handle.net/2031/5768

Title: Cross platform low bit rate video transmission and quality prediction
Other Titles: Kua ping tai di ma lü shi pin tong xin he shi pin zhi liang yu ce
跨平臺低碼率視頻通信和視頻質量預測
Authors: Du, Yufei (杜宇飛)
Department: Department of Computer Science
Degree: Master of Philosophy
Issue Date: 2009
Publisher: City University of Hong Kong
Subjects: Video telephone.
Mobile communication systems.
Notes: CityU Call Number: TK6505 .D8 2009
x, 103 leaves : ill. 30 cm.
Thesis (M.Phil.)--City University of Hong Kong, 2009.
Includes bibliographical references (leaves 100-103)
Type: thesis
Abstract: The unification of 3G and video sensor of real-time video streaming is significance for the design of the supporting architecture that transmits the multimedia streams between video sensor and 3G networks (UMTS). Video sensors can provide important visual information in a number of applications including health care, environmental monitoring, emergency response and security surveillance. The quality of these videos which especially transmit to a mobile user is difficult to estimate and predict because many factors will interfere the result. For example, encode, decode and transmission lose will dramatically reduce the video quality. Predicting the video quality enables many uses, e.g. choosing the best video quality channel when the users transmit the video. This work contributes to integrate 3G video transmission in 3G-324M protocol stack with video sensors and video quality measurement and predication, especial, we have designed an algorithm to predict the received video quality base on the information of original video type and channel status information. It can improve the performance of H.245 protocol under 3G-324M in terms of video transmission. Chapter 2 studies H.324M with some basic knowledge which forms 3G-324M protocol stack for low bit rate multimedia communication in 3G circuit switched network. We present an efficient implementation of controlling and data multiplexing protocols H.245 and H.223, target for multiple video users in both 3G and Internet environments. With testing on the prototype system, the performance of our protocol realized in our system is satisfactory and stable. Chapter 3 and 4 focus on the design and implementation issues of 3G gateway development. We investigate the efficiency, interoperability, scalability and interconnectivity problems of the efficient 3G gateway. We adopt an efficient model for speeding up the process of encoding the ITU-T H.245 ASN.1 formatted messages into binary stream. In this method, H.245 sets of messages are compiled, used and stored in a table. For encoding further incoming messages, the table is queried and the matching result is used to replace the recompiling messages again and again. Experimental results show that our algorithm greatly improves the performance of call handling in a 3G gateway. Also our method is tested to be fully compatible with existing 3G protocols. Chapter 5 contributes to the effort by introducing a new architecture for video transmission from a video sensor to the UMTS terminal at a low cost and power. The purpose is that the video sensor can communicate with UMTS for the best utilization of UMTS coverage. The integration of the mobile network infrastructure and sensor ad hoc networks can reduce the cost of building new infrastructure and enable the large scale deployment of sensor networks. Chapter 6 presents our real implementation work for the video quality measurements and predictions in the low bit rate video. The Perceptual Evaluation of Video Quality (PEVQ) is introduced and exploited for video quality prediction. The objective of this chapter is to find the relationship of the spatiotemporal dynamics of the content and network status to deduce perceptual quality of a H.263 based encoded video. Once this relationship has been digged up, it is shown that it can be further exploited towards proposing a technique for video quality prediction at a pre-encoding state.
Online Catalog Link: http://lib.cityu.edu.hk/record=b2375028
Appears in Collections:CS - Master of Philosophy

Files in This Item:

File Description SizeFormat
abstract.html134 BHTMLView/Open
fulltext.html134 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer