Title: Maximum Flows on Wireless Lattice Networks
Speaker: Professor Wing Wong
Date: 3 July 2007
Time: 9:15-10:00
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
In this talk we will investigate some fundamental properties of traffic flows on a special class of wireless mesh networks, known as the wireless lattice networks. The focus will be on addressing the maximum concurrent flow problem on such networks.

Biography
Wing Shing Wong joined the Chinese University of Hong Kong since 1992 and is now a Professor of Information Engineering. He was the Chairman of the Department of Information Engineering from 1995 to 2003 and is currently serving as the Dean of the Graduate School. He served as the Science Advisor at the Innovation and Technology Commission of the HKSAR Government from 2003 to 2005. He is a Co-Editor-in-Chief of the Communications in Information and Systems. He has participated in a variety of research areas including mobile communication systems, nonlinear filtering, search engine, and estimation and control of finite communication bandwidth systems.

Title: Wireless Technologies and Applied R&D
Speaker: Dr. Chihlin I
Vice President and Group Director
Communications Technologies Group
ASTRI
Date: 3 July 2007
Time: 10:00-10:45
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
Emerging wireless technologies such as 3G/4G digital cellular, WiMAX, next generation WiFi, DVB and UWB are ushering the 21st century into a second information age. It is in effect a catalyst to converge Internet, consumer electronics, and the mobile communications industry, thus engendering a whole host of next generation products and applications. Further more, multimode multifunction, high spectral efficiency, and miniaturized design has been the trends and areas the industrial players strive to differentiate their respective uniqueness and value-adds.
This talk will address the state-of-art applied R&D in various wireless communications technological segments, which presents abundant opportunities for innovative Intellectual Property Rights (IPR) creations. The identified areas are Multifunction System Technologies, High Spectral Efficiency Radio Access Technologies, and Miniaturization Technologies. Key technical substance, IPR opportunities, and practical applications of each area will be illustrated. Systems, components, material or process technologies will be explored as appropriate. In addition, an integration of the above technologies for eHome-Health-Care applications is exhibited to demonstrate how wireless communications technologies may benefit our daily life even further.
The Applied Science and Technology Research Institute (ASTRI), as the host of Hong Kong ICT R&D Center, has recognized the aforesaid technology trends and opportunities. It has been focusing on developing self-owned IPRs in these areas for the local wireless industry in Hong Kong, Pear River Delta, and China to help strengthen and upgrade the region’s applied technology.

Biography
Dr. Chih-Lin I, the Vice President and Group Director of Communications Technologies Group, joined ASTRI in 2005 and is leading communication technology R&D programs and teams. After receiving her Ph.D. from the Electrical Engineering Department at Stanford University, Dr. I joined AT&T Bell Labs Research in 1987 and became the Director of Wireless and Access Technologies and infrastructure in 1996, reported directly to the President of IP Technology at AT&T Labs and the Co-President of AT&T Labs. She received the Distinguished Alumni Award of NCTU in 2002. From 2002 to 2005, she was with ITRI and leading Taiwan’s Wireless Technology R&D Program, Communications Industry Development Promotion Program, International Collaborations of the Taiwan Telecommunications Program, etc. She was simultaneously a Professor at NCTU and an Adjunct Professor at NTU.
Dr. I has been playing international leadership positions including Board of Governors, IEEE Communications Society (ComSoc); Director, IEEE ComSoc Meetings & Conferences; Chair, Steering Committee of IEEE Wireless Communications and Networking (WCNC); Chair, Advisory Board of Wireless & Optical Communications (WOCC); Chair, TIA 3G Wireless Task Group II; Chair, Award Committee of IEEE CNSC; General Chair, IEEE WCNC 2007 in Hong Kong.

Title: Emergent Applications of “Statistical Models” in Media Computing
Speaker: Professor Horace Ip
Professor (Chair), Department of Computer Science,
Director, Centre for Innovative Applications of Internet and Multimedia Technologies (AIMtech Centre)
City University of Hong Kong
Date: 3 July 2007
Time: 11:00-11:45
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
Within the context of pattern recognition, statistical models help to capture and characterize ensemble behavior, variability, and the relationships of objects in the same class, as well as their attributes. “Statistical models”, either in functional or geometrical forms, have been increasingly applied in the analysis and understanding of 2D and 3D media content. In this talk, we will give examples of constructing such “models” for learning the semantic context of 2D images that serves to narrow the semantic gap in content-based retrieval, and for reconstructing complex and highly deformable 3D objects in biomedicine. In particular, we will highlight some open problems and practical issues as well as potential applications of such techniques in the development of immersive interactive media applications.

Biography
Professor Horace H. S. Ip received his B.Sc. (Hons) and Ph.D. degrees from University College London (UCL), United Kingdom, in 1980 and 1983 respectively. Presently, he is the Chair Professor in the Department of Computer Science and the founding director of the AIMtech Centre at City University of Hong Kong. His research interests include image analysis, pattern recognition, interactive multimedia systems and IT in education. Prof. Ip serves on the Editorial Board of international journals including: IEEE Transactions of Multimedia, Pattern Recognition, the International Journal of Multimedia Tools and Applications, and he is the Area Editor for Asia of Pattern Recognition Letters.

Title: Quality of Experience in network analysis and design: the case of the PSQA project
Speaker: Dr. Gerardo Rubino
Senior Researcher
INRIA
Date: 3 July 2007
Time: 11:45 - 12:30
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
The ultimate goal when designing an application running on top of a communication network is the satisfaction of the final user. For a multimedia application, this can be instantiated into obtaining a high enough Quality of Experience (QoE). We think that the main component of QoE is the perceived quality, that is, the quality of the multimedia flow as seen by the human user, a clearly subjective concept.

This talk describes the PSQA project and its implementation through several PhD thesis. PSQA stands for Pseudo-Subjective Quality Assessment. It is a technology able to provide a numerical (that is, quantitative) and accurate estimation of the quality of a video, audio or multimedia flow, either one-way (for instance, video streaming) or interactive (for instance, IP telephony), as perceived by the end user, automatically and in real time if necessary. The talk will describe the main components of the technology and its main potential applications, the past and present research activities around it, and some of the perspectives concerning our future efforts. We will also underline some associated open problems, both concerning the theoretical foundations and some important practical aspects.

Biography
G. Rubino is a senior researcher at INRIA. It is the head of the Dionysos team at the Rennes Unit of INRIA, a research group specialized in networking analysis (through stochastic modeling) and design. His main research areas are in applied probability models, and in quality evaluation (QoS, QoE). He has held professor positions at the ENST Bretagne engineering school, Brittany, France, and at the university of the Republic, Montevideo, Uruguay. He presently belongs to the IFIP WG 7.3 (Computer System Modeling and Performance Evaluation).

Title: What are the right problems for Hidden Markov Models to solve?
Speaker: Professor Frank K. Soong
Principal Researcher and Manager
Speech Group
Microsoft Reserch Asia
Date: 4 July 2007
Time: 9:00 - 9:45
Vuene: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
Hidden Markov Model, after its debut for speech recognition in the seventies, has been steadily developed into a full-fledged jack-knife for statistical modeling and found applications wider than speech recognition only. In this talk, some case studies will be presented to show what seem to be the right problems for HMM to solve and what may. No attempt are made to do an exhaustive review of all possible scenarios. The cases to be analyzed are speech recognition and synthesis, speaker recognition, speech enhancement, handwritten character recognition and synthesis, music melody tracking and recognition.

Biography
Frank Soong is currently a Principal Researcher and Manager of the Speech Research Group at Microsoft Research Asia (MSRA), Beijing, China. His Speech Group works on fundamental research and its practical applications, including: automatic speech recognition (ASR), text-to-speech synthesis (TTS) and audio information management and extraction (AIME).
He spent more than 25 years in speech research. Before joining MSRA, he was with the Spoken Language Translation Research Labs, ATR, Japan, as an invited researcher for two years from 2002 to 2004. Prior to that, he worked at Bell Labs Research for 20 years since 1982, first as a Member of Technical Staff (MTS) and later, Distinguished Member of Technical Staff (DMTS). At Bell Labs, his research area covers: speech analysis and synthesis; high quality, medium-to-low bit rate speech coding; automatic speech and speaker recognition; acoustic echo cancellation and hands-free voice interface; noise robust spectral features for speech and speaker recognition; efficient search algorithms; discriminative HMM training; etc. In 1987-88, he was an invited researcher at NTT Electro-Communication Research Labs (ECL), Musashino, Tokyo, Japan, under an exchange research program between NTT and AT&T Bell Labs. At Bell Labs, he led an award-winning team to transfer research technologies to voice-activated dialing mobile phone product (AT&T Consumer Product 3150 and 3050) which has been rated by the Mobile Office Magazine (Apr. 1993) as the “outstandingly the best”. He was a co-recipient of the Bell Labs President Gold Award for developing the Bell Labs Automatic Speech Recognition (BLASR) software package.
He was on the Speech Technical Committee, IEEE Signal Processing Society and served as an Associate Editor of the Speech and Audio Transactions. He co-chaired the IEEE International Arden House Workshop, a predecessor of the IEEE International Automatic Speech Recognition and Understanding (ASRU) workshop.
He published extensively and published more than 100 papers. He gave a tutorial in ICSLP2004, Jeju, Korea, on Adaptation of Acoustic HMM. He co-edited the widely used reference book, Automatic Speech and Speech Recognition- Advanced Topics, Kluwer, 1996.
He is the co-Director of the MSRA-CUHK Joint Research Lab and a visiting professor of the Chinese University of Hong Kong (CUHK) and a few other top-rated universities in China.

Title: Software Challenges for Multicore Architectures
Speaker: Professor Wenguang Chen
Associate Professor
Department of Computer Science & Technology
Tsinghua University
Date: 4 July 2007
Time: 9:45-10:30
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
Multi-core chip architecture is becoming the mainstream solution for next generation microprocessor chips. Multi-core architectures present new opportunities as well as challenges to software that will run on computer systems built up on these multi-core chips. In this talk, we categorize the mainstream applications and point out the main challenge is to provide user-friendly paralell programming language and environment to desktop/embedded system developers. We also discuss some key techniques such as automatic parallelization and parallel debugging.

Biography
Wenguang Chen received the B.S. and Ph.D. degrees in computer science from Tsinghua University in 1995 and 2000 respectively. He was the CTO of Opportunity International Inc. from 2000-2002. Since January 2003, he joined Tsinghua Univeristy. He is now an associate professor and associate head of Department of Computer Science and Technology, Tsinghua University. His research interest is in compiler, programming model, computer system performance analysis and bioinformatics.

Title: Compression Techniques for Evolving Networks
Speaker: Professor Hong-goo Kang, Yonsei University
Date:
Time:
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
Past, current, and future of multimedia compression techniques are discussed in this talk. The efficiency of the compression techniques has been significantly improved for past 30 years. In case of speech and audio signal, for example, transparent quality can be obtained if we use only around 10% to the digitized or pulse code modulated (PCM) input signal. Video broadcasting or streaming services would not be also successful without adopting compression techniques.
As the channel bandwidth of network system is increased, however, a question arises whether we still need compression techniques for next generation communication/broadcasting networks. The available channel bandwidth for each customer increases from several kilo bits per second (kbps) to several millions bits per second (mbps) thanks to xDSL, cablemodem, and 3G/4G wireless networks, which means that no compression might be needed to communicate or to broadcast multimedia signals. It is partly true in some situations, but not in general. The demands for high quality and premium service require large amount of additional information, which means that novel compression paradigms would be still needed.
In this talk, after briefly reviewing the history of compression techniques and current state-of-the-art techniques, I will introduce on-going research trends relating to international standardization activities. Finally, I would like to share and discuss with audience about research directions.

Biography
Hong-Goo Kang received B.S., M.S., and Ph.D. degrees in Electronic Engineering from Yonsei University, Seoul, Korea, in 1989, 1991, and 1995, respectively. From April 1996 to August 2002, he served on the staff of the Speech and Image Processing Services Laboratories of AT&T Labs-Research, Florham Park, New Jersey, as a Senior Technical Staff Member working on low bit-rate speech coding and single channel speech enhancement. In September 2002, he joined the Department of Electrical and Electronic Engineering, Yonsei University, where he is currently an Associate Professor. His research interests include speech/audio signal processing, adaptive digital signal processing, and real-time systems. He is serving an associate editor of IEEE Transactions on Speech, Audio, and Language Processing.

Title: Self-Healing Routing
Speaker:Professor Mingwei Xu
Professor
Department of Computer Science & Technology
Tsinghua University
Date: 4 July 2006
Time: 11:30-12:15
Venue: Cho Yiu Hall, The Chinese University of Hong Kong

Abstract
In this talk, the self-healing ability of current routing protocols is evaluated, and a simple model is introduced. Then some ideas about self-healing IGP, self-healing BGP and topology design are given.

Biography
Mingwei Xu received the B.S. and Ph.D. degrees in computer science from Tsinghua University in 1994 and 1998 respectively. Now he is a professor of Department of Computer Science and Technology, Tsinghua University.