2009-10 Final Year Project List

Project Title	Descriptions	Prerequisites	No. of students	Supervisor(s)/ Contact method	Allocation results
Self-proposed Project			4	Dr  Chan Ho Leung (by appointment)	Hong Feng (fhong)
Facial Expression synthesis	The project is to animate different facial expressions, namely, fear, surprise, sadness, anger, disgust and joy, based on real world images.		2	Dr  Chan K P (10:30-11:30am.)	Chan Yat Cheung
Feature Extraction for Facial Expression Recognition	Facial expression broadly includes: fear, surprise, sadness, anger, disgust, joy. The first thing to do for recognition is feature extraction. Traditionally, features based on action units are being used. In this project, a real-time feature extraction system is to be implemented, that will obtain an image from the web-camera of a notebook computer, and then extract features. They are then fed to a recognition system.		3	Dr  Chan K P (10:30-11:30am.)	Lee Ka Yin Li Wing Yi Yeung Cho Yi
Self-proposed Project			4	Dr  Chan Hubert (by appointment)
A Simulator of Uncertain Location Data	A location-based service is a new and important application. For example, a mobile phone user may be interested to find out which restaurant is closest to him. However, uncertainty is prevalent in these applications due to the accuracy limitation of mobile devices. In this project, we would like to design a computer simulator, which generates location data with uncertainty models, in order to simulate the objects? movement with various uncertainty factors (e.g., measurement, sampling, and transmission). For the purpose of performing scientific experiments and evaluations, such a simulator is highly desirable. The students involved will need to have a good background of probability and statistics, high research potential and eagerness to learn. They will also have a hands-on experience in designing a computer simulator.	C++, Java, some background in probability and statistics	2	Dr  Cheng Reynold (by appointment, on or before 22 June)
Energy-aware location-based services	In this project, we would like to develop a system for providing high-quality services to users. The system will be used to facilitate continuous Location-Based Services (LBS) e.g., which friends are within 5 miles from my location in the next hour? This requires mobile users (e.g., GPS-enabled devices) to continuously report and update their current locations. However, this requires battery power and network bandwidth, which are precious resources in mobile devices. Hence, the system should be able to balance the power consumption and service quality. The students involved will have to develop a backend database (which stores location and other information) to support the service. Because of the uncertainty of the location data obtained (due to limited battery power and network bandwidth), the student may need to use the uncertain database to handle the storage of location data. Moreover, the student will have the opportunity to learn and use a mobile device programming language (e.g., Java ME). The student will need to be competent in programming, software engineering skills. Able students may also have opportunity to involve in exciting research works in this area.	C++, Java, Java ME (recommended), probability/statistics background preferred	2	Dr  Cheng Reynold (by appointment (on or before 22 Jun))	Chan Hao (hchan2) Hui Yan Ting Lau Chi Ho Li Man Ho (mhli)
Energy-aware location-based services (Group 2)	In this project, we would like to develop a system for providing high-quality services to users. The system will be used to facilitate continuous Location-Based Services (LBS) e.g., which friends are within 5 miles from my location in the next hour? This requires mobile users (e.g., GPS-enabled devices) to continuously report and update their current locations. However, this requires battery power and network bandwidth, which are precious resources in mobile devices. Hence, the system should be able to balance the power consumption and service quality. The students involved will have to develop a backend database (which stores location and other information) to support the service. Because of the uncertainty of the location data obtained (due to limited battery power and network bandwidth), the student may need to use the uncertain database to handle the storage of location data. Moreover, the student will have the opportunity to learn and use a mobile device programming language (e.g., Java ME). The student will need to be competent in programming, software engineering skills. Able students may also have opportunity to involve in exciting research works in this area.	C++, Java, Java ME (recommended), probability/statistics background preferred	2	Dr  Cheng Reynold (by appointment (on or before 22 Jun) )
Storage and query of Wikipedia data	Wikipedia is a flagship Web 2.0 application, which has defined a huge collection of concepts in a very organized fashion. Although the contents on Wikipedia are more organized than those on traditional World Wide Web, Wikipedia is still mainly formatted for comprehension by human readers. Some Wikipedia data such as those in the InfoBox can be regarded as RDF data. Since the amount of such data in the Wikipedia is huge, designing a good storage scheme that is space efficient and query efficient is a challenge. In this project, students would be required to experiment with several proposed storage schemes such as triple store, column based table, etc., to design a good storage strategy. A query interface will need to be developed to support a prototype on such as a system.	Java programming	2	Prof  Cheung David (Please talk to my secretary Karen)
[Industrial Project] Enhancing the CEAdmin data management system	Industrial Project Please contact Dr. Anthony Tam for more details and arrangement. Company: Macquarie Securities Ltd. Description: CEAdmin is a data management tool written in C# with WPF, WCF technologies that administers static data residing on a SQL Server database shared by multiple applications. This project will focus on expansion of current functionalities to handle additional business requirements on People, Region and Sector data management. A successful implementation requires attention to the following area: * User interface design * Web services design * Understanding of current data model * Application performance * Application security and control	Database development and administration	2	Prof  Cheung David (By appointment)	Ho Chong Fan Ho King Yiu Lam Shan Shan Wong Kam Ming
Graph searching	Suppose that some intruders hide on vertices or along edges of the graph. We want to find the minimum number of searchers required to capture all the intruders hinding in the graph. Note that the intruders can move along edges and vertices in the graph.	good at and interest in Algorithms	1	Prof  Chin Francis (contact 28592178)	Chan Kit Yee (kychan2)
Research on online call control system			1	Prof  Chin Francis (-)	Lam Ka Cheong
Sequence alignment	Although many algorithms have been proposed for multiple sequence alignment, it remains difficult to obtain accurate alignments, especially when the sequences are very long, such as DNA sequnces. Common techniques to improve alignment accuracy include performing iterative refinements after the initial alignment is constructed using consistency-based pairwise alignments in progressive approaches and incorporating structural alignments. The objective of this project is to devise some effective algorithms for multiple sequence alignment.	interested in algorithm studies	1	Prof  Chin Francis (contact 28592178)
Virtual Machine			4	Dr  Chong C F (by appointment)	Chan Nok Ki
Tracing in the eMule P2P network (Group 1)	eMule has been used extensively in China for file sharing. Students are required to understand the eMule protocol and develop automated tools to monitor search efficiently in the eMule network.		2	Dr  Chow K P (make appointments by email)	Tsang Shun Lun Wan Kit Man (kmwan)
Tracing in the eMule P2P network (Group 2)	eMule has been used extensively in China for file sharing. Students are required to understand the eMule protocol and develop automated tools to monitor search efficiently in the eMule network.		2	Dr  Chow K P (make appointments by email )	Wu Chenhao Zhang Xiang
Vista Forensics Tool - edb searching	Windows uses an indexing system to catalogue all the files and folders on the hard disk. The index is stored in the edb file. Students are required to decode and analyze the edb file in the Vista environment, and then build tools to extract and search the data in the file.		1	Dr  Chow K P (make appointments by email)
Vista Forensics Tool - Link file analyzer	Link files are used extensively in Windows environment. Students are required to study the structure and behavior of the Windows link files and then implement tools to analyze and interpret the link files.		1	Dr  Chow K P (make appointments by email)	Fung Kai Chiu
Free SMS (Group 1)	Short Message Service (SMS) is a communication service standardized in the GSM mobile communication system allowing the interchange of short text messages between mobile telephone devices. However, sending SMS is usually charged on a per message basis, and sometimes the charge is higher when SMS has to be sent across different mobile network providers. In this project, we would like to explore the possbility of tunneling SMS messages through 3G IP networks, to come up with a "Free SMS" delivery solution (assuming unlimited 3G data tranmission plan). The SMS tunneling should be transparent to users, and the solution is expected to work on Windows Mobile platforms.	Good programming in C/C++, and have interests in Windows Mobile Programming	2	Dr  Chung Ronald (By Appointment)	Hung Ka Lai Ng Sin Ling
Free SMS (Group 2)	Short Message Service (SMS) is a communication service standardized in the GSM mobile communication system allowing the interchange of short text messages between mobile telephone devices. However, sending SMS is usually charged on a per message basis, and sometimes the charge is higher when SMS has to be sent across different mobile network providers. In this project, we would like to explore the possbility of tunneling SMS messages through 3G IP networks, to come up with a "Free SMS" delivery solution (assuming unlimited 3G data tranmission plan). The SMS tunneling should be transparent to users, and the solution is expected to work on Windows Mobile platforms.	Good programming in C/C++, and have interests in Windows Mobile Programming	2	Dr  Chung Ronald (By Appointment)	Chan Man Chi Chan Yat Sing
GPGPU-based H.264 Video Decoding	The objective of this project is to develop a flexible framework and toolset for utilizing the General Purpose Graphics Processing Unit (GPGPU) in high-definition video decoding and post-processing. The GPGPU coprocessor approach is expected to provide better handling in multiple video streams decoding and post-processing, and to allow more flexibility in scheduling decoding tasks to minimize end-to-end video delivery latency which is crucial for application, such as remote monitoring in digital surveillance system, that requires instant real-time video distribution over IP. The GPGPU-based video decoding and post-processing framework resulted from this project is expected to help surveillance industry to bring H.264 as well as HDTV technologies into their existing solutions.	Good in C/C++ programming, and have interests in video processing.	2	Dr  Chung Ronald (By Appointment)	Siu Chi Shing Tang Benjamin Sui Lueng (bsltang)
Analyzing Stock Quotes using Data Mining Techniques (Group 1)	One famous approach to stock trend prediction is technical analysis. Technical analysts believe that the stock prices (open, high, low, close, volume) carry significant information on price prediction. With such a large volume of available stock quote data online (e.g., in Yahoo! Finance), we can use data mining techniques (classification, clustering, etc.) to dig out useful information (e.g., frequent patterns) from stock price history. This project aims to develop stock price prediction models based on the information we can dig out.		2	Dr  Ho W S (by appointment)	Tong Chiu Man
Analyzing Stock Quotes using Data Mining Techniques (Group 2)	One famous approach to stock trend prediction is technical analysis. Technical analysts believe that the stock prices (open, high, low, close, volume) carry significant information on price prediction. With such a large volume of available stock quote data online (e.g., in Yahoo! Finance), we can use data mining techniques (classification, clustering, etc.) to dig out useful information (e.g., frequent patterns) from stock price history. This project aims to develop stock price prediction models based on the information we can dig out.		2	Dr  Ho W S (by appointment)	Lo Ioi Ian Luk Chi Wa
Visualization of Common Patterns in Stock Quotes	This project aims to develop a tool for extracting common patterns from time series and displaying them in a summarized form. For example, the system extracts interesting patterns from different stocks and visualizes them. The user can refine the summary and try to extract useful knowledge from the graph.		2	Dr  Ho W S (by appointment)	Duan Lian Li Sulei
Study of Web-based Systems Security Technology (Group 1)	As Internet becomes more popular, there are also a lot of criminal activities involving digital data. Some of them are related to technical weakness of the security design, for example the disclosure of customer data at the server database, USB thumb drive, or more advanced 'SSL session hijacking' (e.g. the e-Gold case) which can by-pass almost all current available authentication technology. However, a lot of those security vulnerabilities can be avoided if proper security measures (e.g. client side re-authentication, server-side authentication, media file encryption, secret sharing, etc) are incorporated. In this project, students are expected to perform design and develop work of prototype systems which can address some common security threats. Last year a similar FYP was purposed to implement a web-based system such that the content in the web pages are encrypted, and will only be seen by providing the correct decryption key in the client side. Some possible project topics are listed here: ======= beginning of list ======= 1. Online Library In the online library, user can borrow books and view it in digital format. It requires a database to store client¡¦s username, book list (i.e. books he/she has borrowed), encrypted password and client¡¦s public key. The books are uploaded by the administrator of the website instead of clients. When the book is uploaded, it is stored in digital format and the pages of the book are stored in the directory which is specific to that book. That means, at server side, there are different directories and for each directory, it stored the digital pages for a particular book. For every registered client, he is given a Java Card. If the user wants to view any books, he/she can borrow through some borrow/return systems and then borrowed books will be added to the user¡¦s book list at server side database. When the user wants to view a particular book, he/she can choose to view which page of that book. Once the page is selected, the encrypted page at the server side will be re-encrypted by another AES key and the AES key will be encrypted by the user¡¦s public key and sent to the client. Then, the Java Applet will help to decrypt the page by using Java Card¡¦s key and show it on the browser. (This will be an improvement of last year¡¦s project. Focus is more on web design) Some administrative problem such as some users losing their Java Card, have to be addressed. 2. Online school or tutorial materials There are many tutorial schools that will upload the tutorial or school materials to the web for e-learning or supplementary materials of the class. Since these materials can only be seen by the students, last year¡¦s FYP system (an encrypted file sharing system as mentioned before) could be used to achieve this security purpose. It can be used for preventing course materials being shown to others. Firstly, for the database structure, each student¡¦s name, password and his/her chosen courses are stored in the database. Furthermore, different courses¡¦ materials will be stored in different directories. Therefore, student can view the materials of the chosen courses. The upload process will be held by the administrator of the course. The course materials will be encrypted by an AES key and the AES key is encrypted by server¡¦s public key and stored at server side. If a student wants to view the materials in the chosen course, the server will re-encrypt the materials with another AES key and the AES key will be encrypted by the student¡¦s public key, so that the student could download, retrieves the AES key, decrypts the course materials and view it in the browser. A variation will be that each student can take online tests, and the test results will be either stored in the web server in encrypted form, or stored at the Java Card at the client side. (This will be an improvement of last year¡¦s project. Focus is more on cryptographic operation design.) 3. Java Card Communication Systems Create a Java Card-based secret communication system such that two communication parties will exchange secret messages in a convenient manner. A variation of this project will be to build a similar system for a group of more than two communicating parties. (Focus is more on the cryptographic design and system design.) 4. Technology infrastructure development. Improve the last year¡¦s system by one of the following infrastructure development. -?Ü Improve decryption speed of the files by design/implementation of the ¡¥multiple level key¡¦ structure. -?Ü Providing a Time-sync service with Java Card, and time-based decryption system. -?Ü Adding One-time-password, or count-based authentication system -?Ü Adding a delegation sub-system such that one Java Card owner can delegate the decryption right of a document to another -?Ü Adding a delegation sub-system such that a Java Card will delegate the file display right to a SD card. The SD card will contain a fixed number of keys. Each key will be able to see the file one time. The Java Card owner can revoke the decryption right of the SD card owner. -?Ü Implementing cryptographic primitives such as Merkle hash tree, or elliptic curve cryptography to the existing system. (Focus is more on understanding and implementation of cryptography/security technology) 5. Advanced Research Topics A student will carry out any other research and development topics after discussion with the supervisor. ======== end of list ======== The above is just a general description. The project students will have discussion with the supervisor (hui@cs.hku.hk) to limit their focus to one aspect, depending on the interest/ability of the students. Students are expected to work independently for one topic, or to work in a team of two persons with clear division of labor. It is also possible that more than one group of the students are accepted to work on different aspects of this project. Therefore, ¡§Students can apply for this project INDIVIDUALLY, or in group of size 2.¡¨ The supervisor will arrange with the students accordingly. The most important attribute for working on this project is the interests in learning topics in computer security and cryptography. Students will be provided with some already implemented tools, a copy of last year¡¦s final year project (so instead of building everything from scratch, the students may start with improving the system built by last year¡¦s project. The link http://www.cs.hku.hk/cisc/DrHui/FYP-08-09-info_Rev.zip contains a presentation about last year¡¦s system), some open source packages, hardware tokens (such as Java Cards and secure USB thumb drives), and literatures. Students interested in this project are STRONGLY encouraged to contact the supervisor (hui@cs.hku.hk) for more information.	Interests in computer security and cryptography, and reasonable programming skill.	4	Dr  Hui Lucas (June 22 (Mon) 2-4pm (A demo of last year¡¦s system is also available on June 22), Venue is CYC307.)	Ho Sze Lok (slho)
Study of Web-based Systems Security Technology (Group 2)	As Internet becomes more popular, there are also a lot of criminal activities involving digital data. Some of them are related to technical weakness of the security design, for example the disclosure of customer data at the server database, USB thumb drive, or more advanced 'SSL session hijacking' (e.g. the e-Gold case) which can by-pass almost all current available authentication technology. However, a lot of those security vulnerabilities can be avoided if proper security measures (e.g. client side re-authentication, server-side authentication, media file encryption, secret sharing, etc) are incorporated. In this project, students are expected to perform design and develop work of prototype systems which can address some common security threats. Last year a similar FYP was purposed to implement a web-based system such that the content in the web pages are encrypted, and will only be seen by providing the correct decryption key in the client side. Some possible project topics are listed here: ======= beginning of list ======= 1. Online Library In the online library, user can borrow books and view it in digital format. It requires a database to store client¡¦s username, book list (i.e. books he/she has borrowed), encrypted password and client¡¦s public key. The books are uploaded by the administrator of the website instead of clients. When the book is uploaded, it is stored in digital format and the pages of the book are stored in the directory which is specific to that book. That means, at server side, there are different directories and for each directory, it stored the digital pages for a particular book. For every registered client, he is given a Java Card. If the user wants to view any books, he/she can borrow through some borrow/return systems and then borrowed books will be added to the user¡¦s book list at server side database. When the user wants to view a particular book, he/she can choose to view which page of that book. Once the page is selected, the encrypted page at the server side will be re-encrypted by another AES key and the AES key will be encrypted by the user¡¦s public key and sent to the client. Then, the Java Applet will help to decrypt the page by using Java Card¡¦s key and show it on the browser. (This will be an improvement of last year¡¦s project. Focus is more on web design) Some administrative problem such as some users losing their Java Card, have to be addressed. 2. Online school or tutorial materials There are many tutorial schools that will upload the tutorial or school materials to the web for e-learning or supplementary materials of the class. Since these materials can only be seen by the students, last year¡¦s FYP system (an encrypted file sharing system as mentioned before) could be used to achieve this security purpose. It can be used for preventing course materials being shown to others. Firstly, for the database structure, each student¡¦s name, password and his/her chosen courses are stored in the database. Furthermore, different courses¡¦ materials will be stored in different directories. Therefore, student can view the materials of the chosen courses. The upload process will be held by the administrator of the course. The course materials will be encrypted by an AES key and the AES key is encrypted by server¡¦s public key and stored at server side. If a student wants to view the materials in the chosen course, the server will re-encrypt the materials with another AES key and the AES key will be encrypted by the student¡¦s public key, so that the student could download, retrieves the AES key, decrypts the course materials and view it in the browser. A variation will be that each student can take online tests, and the test results will be either stored in the web server in encrypted form, or stored at the Java Card at the client side. (This will be an improvement of last year¡¦s project. Focus is more on cryptographic operation design.) 3. Java Card Communication Systems Create a Java Card-based secret communication system such that two communication parties will exchange secret messages in a convenient manner. A variation of this project will be to build a similar system for a group of more than two communicating parties. (Focus is more on the cryptographic design and system design.) 4. Technology infrastructure development. Improve the last year¡¦s system by one of the following infrastructure development. -?Ü Improve decryption speed of the files by design/implementation of the ¡¥multiple level key¡¦ structure. -?Ü Providing a Time-sync service with Java Card, and time-based decryption system. -?Ü Adding One-time-password, or count-based authentication system -?Ü Adding a delegation sub-system such that one Java Card owner can delegate the decryption right of a document to another -?Ü Adding a delegation sub-system such that a Java Card will delegate the file display right to a SD card. The SD card will contain a fixed number of keys. Each key will be able to see the file one time. The Java Card owner can revoke the decryption right of the SD card owner. -?Ü Implementing cryptographic primitives such as Merkle hash tree, or elliptic curve cryptography to the existing system. (Focus is more on understanding and implementation of cryptography/security technology) 5. Advanced Research Topics A student will carry out any other research and development topics after discussion with the supervisor. ======== end of list ======== The above is just a general description. The project students will have discussion with the supervisor (hui@cs.hku.hk) to limit their focus to one aspect, depending on the interest/ability of the students. Students are expected to work independently for one topic, or to work in a team of two persons with clear division of labor. It is also possible that more than one group of the students are accepted to work on different aspects of this project. Therefore, ¡§Students can apply for this project INDIVIDUALLY, or in group of size 2.¡¨ The supervisor will arrange with the students accordingly. The most important attribute for working on this project is the interests in learning topics in computer security and cryptography. Students will be provided with some already implemented tools, a copy of last year¡¦s final year project (so instead of building everything from scratch, the students may start with improving the system built by last year¡¦s project. The link http://www.cs.hku.hk/cisc/DrHui/FYP-08-09-info_Rev.zip contains a presentation about last year¡¦s system), some open source packages, hardware tokens (such as Java Cards and secure USB thumb drives), and literatures. Students interested in this project are STRONGLY encouraged to contact the supervisor (hui@cs.hku.hk) for more information.	Interests in computer security and cryptography, and reasonable programming skill.	4	Dr  Hui Lucas (June 22 (Mon) 2-4pm (A demo of last year¡¦s system is also available on June 22), Venue is CYC307.)	Chung Man Ho (mhchung) Tang Chi Chun (cctang) Wat Chun Pang Gilbert (cpgwat) Yeung Siu Wai
Tags recommendation in social networks	In many social networks like flickr, objects, such as images, are tagged by different users using different keywords. We evaluate how this three-dimensional interaction, namely, objects, tags, and users improve information retrieval as compared to the traditional two-dimensional model, which includes only objects and tags.		2	Dr  Kao Benjamin (by appointment)
[Industrial Project] iServices	Industrial Project Please contact Dr. Anthony Tam for more details and arrangement. Company: Dah Sing Bank Description: The project is to make use of net meeting technology to design and develop a new sales platform whereby a customer located in a bank branch or at home (over the Internet) can meet with a bank representative in a remote location. The project team designs not only the system itself but also the sales process that utilizes it.		2	Dr  Kao Benjamin (Closed)	Gu Ying Jin Qi Yan Haosu (hsyan)
Check-My-Play! for the piano	Given a score for a piano piece, write a program to capture the playing of the piece and then check it against the score. The output is a "report card" showing the problems and errors of the playing. The simple version will look for only such simple things as whether the correct notes have been played. The more advanced version can cater for the notes' durations, velocities, phrasing, tempo, etc. In theory, the idea can be applied to any instrument, but the piano (I mean the keyboard) is the easiest for the capturing part. If successful, this will be a great tool for students of the piano. A future version can even do "tutoring" instead of just error-checking.		1	Prof  Lau Francis (by appointment)	Cheng Lee
Intelligent graphical user interface	Intelligent systems are best to be coupled with a user interface to allow the intervention of human intelligence. Reversely, graphical user interface needs the assistance of machine intelligence to reduce the amount of user interaction. In this project, we intend to explore design and implementation issues in intelligent graphical user interface research. Two application fields are targeted: cartoon sketching and document layout design where intelligent graphical user interfaces are most needed.		1	Prof  Lau Francis (by appointment)	Chiang Sze Cheuk
Mining aggregative knowledge on the Web	Today's Web is the world's largest knowledge base. However, information is very "scattered", and much of which needs non-trivial human processing efforts. Acquiring aggregative knowledge from the Web is worse, if not impossible. In this project, we would like to devise ways to automatically extract basic knowledge points from webpages. With masses of such points, we might then be able to infer useful aggregative knowledge. We will apply the method to such domains as public health, social opinions, marketing statistics, and social recommendation, etc. We also plan to develop a graphical frontend to intuitively present the summarized aggregative knowledge to various user groups. This project consists of multiple components, including natural language understanding, web mining, visualization, and applied machine learning.		1	Prof  Lau Francis (by appointment)	Cheung Sin Hang
New web-based media or social entertainment platform (Group 1)	We will study and construct prototype website(s) offering new media entertainment or social entertainment options based on Web X.0 techniques and concepts. We will look at what is next to Facebook or YouTube and try to come up with candidate proposals. This project emphasizes commercial promises and prefers students who are business oriented.		3	Prof  Lau Francis (by appointment)	Shing Yi Ting Ng Wai Lun (wlng)
New web-based media or social entertainment platform (Group 2)	We will study and construct prototype website(s) offering new media entertainment or social entertainment options based on Web X.0 techniques and concepts. We will look at what is next to Facebook or YouTube and try to come up with candidate proposals. This project emphasizes commercial promises and prefers students who are business oriented.		3	Prof  Lau Francis (by appointment)	Lau Mei Ying Ng Che Kit Ting Ka Cheuk Wong Yuet Yi
Personalization and Recommender Systems	Details of these projects can be found here: http://www.cs.hku.hk/~fcmlau/recommender.rar These are group projects, with 2-3 students per group.		2	Prof  Lau Francis (By appointment)
Child care and learning system	To design and implement an integrated system for parent to help caring of children (tracking of growth, activities),and learning activities (color, word, music, etc). Include mobile device such as iPhone or Windows Mobile, and a server.	iPhone or WM tool, and Web server	5	Dr  Lau Vincent (via email and after 19 June, 2009)	Chan Wai Man Cheung Wai Fung Chiu Chi Ming Fung Ching Yiu Lau Chun Yin
Chinese character analysis and construction	Computer analysis of the data of component tree of over 10,000 Chinese characters, try to optimize and generate the character shape with similar quality.	C/C++ and graphics	3	Dr  Lau Vincent (via email and after 19 June 2009)	Shum Helen Jing Yan Ngan Man Chun
Mobile universal remote control system	Give user an easy to use user interface for customization of button and device. May base on open source project www.lirc.org, which contains different remote control IR code. Hardware devices are provided.	C++, pHp or Flash	3	Dr  Lau Vincent (via email and after 19 June 2009)	Cheng Wing Hong Cheung Lok Kan Sit Cheong Tik
Smart PC Console	Develop a smart PC console with internet connectivity for home applications		3	Dr  Lau Vincent (by appointment)	Kwan Kok Sum Lam Chun Kin Tong Kin Man
Fair Assignment Problems	This project is about implementing and testing methods for generating a fair matching between two sets. For example consider a set of public houses that should be allocated to a set of families. Each family sets a number of criteria for selection (e.g., desired number of rooms, distance to city centre, etc.). The criteria could be constraints (e.g., I cannot live in a house with only one room) or preferences(e.g., I prefer a large house, but I don't mind a smaller one). The houses carry features that are considered in these criteria. For example, a home could be an apartment on the 10th floor, facing east, with 2 bedrooms, total size 650 sq. f., and in Wan Chai district. Our objective is to find a fair matching between the houses and the families that request them. Fairness can be defined according to different optimization functions (e.g., minimize the total number of violated constraints and then maximize an aggregate in the remaining preferences). The ultimate goal of the project is to verify the appropriateness of different objective functions in practice.	none	3	Dr  Mamoulis Nikolaos (after 17 June, 2009)	Lam Tsz Fung To Sing Yin Kenneth Wong Chi Hin
Time-travel Text Analysis	This project is about implementing a number of operations for analyzing the results of keyword queries (e.g., Google queries) over time. This type of search is mainly suited for news articles, which are relevant to only a short period of time. For example, consider the query "swine flu", the time interval 2009, and the time granularity "week". For each week of 2009, the goal of the search is to show the most relevant articles to the query. Please check http://newstimeline.googlelabs.com/ for an example of this type of search already supported by Google. The goal of this project is to extend this type of queries to more complex querying tasks. For example,given the articles that are results of a keyword query, what are the most interesting phrases of them for each granule of the time period (e.g., each week)? Can we support interactive operations from OLAP, such as roll-up and drill-down in the level of analysis? The students will be given a large document corpus (NY times articles) to use as input document collection and will be required to create an advanced, highly interactive user interface for such text analysis.	Good knowledge of UI development, good programming skills	3	Dr  Mamoulis Nikolaos (after June 17, 2009)	Cheung Ching Lung
Chinese search engine for the Community Legal Information Centre (CLIC)	CLIC (www.hkclic.org) is a bilingual database maintained by the Department of Computer Science offering free legal information to Internet users with no legal background. The database comprises of documents written in English and Chinese in the form of FAQs on 21 legal topics of common interest to the general public. The database currently has a search engine for English documents that accepts queries using English keywords. The objective of this project is to design and implement a search engine for Chinese documents that accepts queries using Chinese keywords.		1	Dr  Pun K H (By appointment)
Technology and copyright			1	Dr  Pun K H (By appointment)	Chow Ho Yin
Technology and privacy (Group 1)	To conduct a survey on existing computer technologies which have privacy implications and the potential liabilities arising from such technologies under the existing privacy law of Hong Kong. The survey will look at both the technological and the legal aspects of such technologies.	CSIS 0311	1	Dr  Pun K H (By appointment)	Siu Ka Wai
Technology and privacy (Group 2)	To conduct a survey on existing computer technologies which have privacy implications and the potential liabilities arising from such technologies under the existing privacy law of Hong Kong. The survey will look at both the technological and the legal aspects of such technologies.	CSIS 0311	1	Dr  Pun K H (By appointment)	Tse Wai Hang
Parallel programming on multicore and manycore	Learning how to write parallel program is challenging for undergraduate CS students. However, learning how to write highly efficient parallel program on various multicore (or even manycore) systems is even more challenging. Due to the rapid advances in processor technologies, we now have a vast diversity of available architectures that support various forms of parallelism, such multicore, multiprocessor, GPGPU, .., etc. A parallel implementation may run efficiently on a particular architecture, but this does not guarantee to be run efficiently in other architectures. Efficiency and portability are two conflicting goals. In this project, the student will focus on a particular class of applications/algorithms and try to investigate on how to design and implementing parallel solutions that can balance between efficiency and portability.	CSIS0230 and thread programming	1	Dr  Tam Anthony (Weekdays)
[Industrial Project] Centralized monitoring system for large clusters of servers	Industrial Project Please contact Dr. Anthony Tam for more details and arrangement. Company: Thomson Reuters Hong Kong Ltd. Title: Centralized monitoring system for large clusters of servers Description: The company delivers real-time data through large number of distribution servers and edge devices installed in data centres located throughout the world. It is crucial to have a centralized monitoring system with easy access and well-designed interface to support operations staff, product managers and capacity managers in their daily work, e.g. to monitor the healthiness of all servers; to keep track of system performance; and to analyze trend graphs for planning any system upgrades or service improvements. In addition to passive monitoring, the system should also provide active notification to company staff via email so that prompt action can be taken in the event of critical problems on the servers.	CSIS0230 and Database development & administration	3	Dr  Tam Anthony (Closed)	Cheung Ching Yu (cycheung) Tan Kok Kong Jonathan Tan Kok Thye Joseph
[Industrial Project] Detection of exceptional behaviour of IT systems based on intelligent analysis of system statistics	Industrial Project Please contact Dr. Anthony Tam for the details and arrangement Company: Thomson Reuters Hong Kong Ltd. Title: Detection of exceptional behaviour of IT systems based on intelligent analysis of system statistics Description: The company runs many business-critical 24x7 server applications. To allow fast resolution of software or hardware problems, each server application records a large amount of statistics on utilisation of the system resources, processor loading of the applications, processing errors detected, invalid input detected, etc. To maintain the promised 99.95% availability, it is important to locate the root cause of a problem as quickly as possible on the basis of system statistics. On the other hand, some unexpected deviation of a metric from its normal trend may also indicate potential issues for a server or its environment. Locating such a deviation may help locate a problem before client service is affected. Each application type, however, can collect hundreds or thousands of different statistics. It is very time consuming to go through all statistics from all machines to locate any abnormality. In addition, pre-defined rule-based analysis can only be effective for known scenarios. An intelligent analysis system is required to reduce the time to locate abnormalities before and after a problem occurs.	CSIS0230	2
[Industrial Project] Multicast network service monitoring application	Industrial Project Please contact Dr. Anthony Tam for the details and arrangement Company: Thomson Reuters Hong Kong Ltd. Title: Multicast network service monitoring application Description: IP Multicast is commonly used for sending information to hundreds or thousands of consumers simultaneously. However, implementing reliable multicast delivery is very complicated when the number of users reaches hundreds or more. Use of unreliable multicast will introduce data loss and corresponding support and troubleshooting concerns. A monitoring application with good user interface is critical to a multicast service over a network with multiple levels of routers or switches between a data transmitter and (hundreds of) consumers. To provide quick response to any support call, the monitoring application based on Microsoft Silverlight is expected to monitor flow of data at multiple points in the network and provide intelligent alert or notification on possible failure points or scenarios.	CSIS0234	3
[Industrial Project] Research and develop a data caching strategy	Industrial Project Please contact Dr. Anthony Tam for more details and arrangement. Company: Macquarie Securities Ltd. Description: Research and develop a data caching strategy for in-house developed applications to support the Research/Publication business unit. An effective data caching mechanism is important for application performance and end user experience. Explore different Microsoft caching technologies on server/client-side, e.g. Microsoft Enterprise Library Caching, SQL merge replication with Compact edition, Velocity, etc. Improvements with large dataset should consider the following factors: * Application start-up performance * Application response (search efficiency, display, etc) * Network bandwidth utilization * Recovery from corrupted data * Server Load (Disk/Memory/CPU) * Development Overhead	Database development and administration	2
Algorithm studies on business problem analysis			1	Dr  Ting H F (-)	Liu Zhe
School Timetable Construction			4	Dr  Ting H F (by appointment)	Choi Wing Kar Chan Ka Chun Kelvin-chem (kckchan)
[Industrial Project] Web based system for managing payment files	Industrial Project Please contact Dr. Anthony Tam for more details and arrangement. Company: The Hong Kong & China Gas Company Ltd. Description: This project is to create a web based system for managing payment files from different financial institutions via email and internet. The system will be developed on .Net and SQL-server. The system is required to: * Automate the transmission of payment record files from and to financial institutions. * Support PGP encryption and decryption of payment record files. * Support automatic archiving and housekeeping of these files. * Provide processing logs/reports for review. * Support audit trail functions. * Support different security access levels to the system. * Provide interface/web services to Job scheduling system. * Provide statistics on system monitoring and control.	.NET application development	2	Dr  Ting H F (Closed)	Cheung Chi Kong (ckcheung) Miu Wa Ying (wymiu)
Development of a multiplayer online game	The objective of this project is to design and implement an interesting and exciting multiplayer online game playing via the Internet. The basic requirement is real-time 3D interactive animation with audio effect. Other optional features include artificial intelligence, collision detection, particle system, etc. The game shall be developed using a 3D modeling software, e.g., Maya and a game engine, e.g., the Microsoft XNA Game Studio. Motion capture may be used for improving the realism of the animation. Participants are expected to take CSIS0271 on computer graphics and CSIS0329 on computer game design and programming in the coming year.		4	Dr  Tsang W W (10:30-11:40am, Monday and Tuesday)	Dai Xin
Towards Improved Techniques in Fault Localization	Program debugging is laborious, and fault localization is recognized as its most difficult component. Popular statistical fault-localization techniques automatically rank the statements or predicates of a faulty program to produce a suspicious list of program entities. Despite their popularity, our initial empirical studies find that there is room for improvement for their effectiveness. We propose a project to study improved techniques for fault localization.		8	Prof  Tse T H (Please feel free to contact me via email at thtse@cs.hku.hk.)	Hui Kin Hing (khhui)
Ad hoc Location Tracking with Mobile Landmarks	Read project descriptions in http://i.cs.hku.hk/~clwang/FYP/FYP-2009-2010.htm		3	Dr  Wang C L (10:00pm June 13 (Saturday), or by appointment)	Kwok Ka Ho Kwok Wan Lai Ip Chun Yin (cyip)
BetterLife 2.0: Personalized Recommendation Service on Cloud	Read project descriptions in http://i.cs.hku.hk/~clwang/FYP/FYP-2009-2010.htm		3	Dr  Wang C L (10:00am, June 13 (Saturday), or by appointment)	Kong Kwai Yee Lo Fung Wong Kwok Kit
Flash-based 3D Visualization Front-end	This project is about internet-based 3D environmental impact assessment (EIA), with which the user can visualize the Hong Kong terrain and water quality variations in 3D within a browser. The realistic 3D terrain and GIS data are generated by a server. The front-end's functions are to composite the terrain with dynamic simulation results in 3D using Adobe Flash's 3D/Animation features and to provide basic interface for navigation and data interrogation functions. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP1 Contact: Prof. Wang (wenping@cs.hku.hk) Dr. Bin Chan (bchan@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	2	Prof  Wang W (Between June 17-June 24, office hour or by appointment or email)
Hierarchical 3D Object Segmentation	The aim of this project is to segment a 3D object into meaningful components. The tetrahedral elements of an input 3D object are clustered hierarchically until certain criteria are satisfied. The main steps of this algorithm follow [1]. Reference: [1] Marco Attene, Michela Mortara, Michela Spagnuolo and Bianca Falcidieno. Hierarchical Convex Approximation of 3D Shapes for Fast Region Selection. Computer Graphics Forum, Vol. 27, No. 5 (SGP'08 Procs.), pp. 1323-1333, 2008 More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP5 Contact: Prof. Wang (wenping@cs.hku.hk) Mr. Dongming Yan (dmyan@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)
High-quality Mesh Generation	The aim of this project is to develop a new method for generating a 2D quality mesh. The input is a 2D polygonal domain and the output is a 2D mesh filling the input polygon with well-shaped triangles. We will use the framework described in [1], however the mesh generation algorithm will be different. The requirement includes basic knowledge about graphics programming, such as OpenGL. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP4 Reference: [1] Jane Tournois, Pierre Alliez and Olivier Devillers. Interleaving Delaunay Refinement and Optimization for 2D Triangle Mesh Generation. 16th International Meshing Roundtable Contact: Prof. Wang (wenping@cs.hku.hk) Mr. Dongming Yan (dmyan@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)	Yu Choi Kuk
Interactive Image Segmentation	The aim of this project is to develop user interface for efficient image segmentation. For a given 2D image, the system should provide functions for the user to draw strokes to specified foreground and background, and uses the lazy snapping technique[1] to achieve high quality segmentation. Extension to medical data, 2D or 3D, is a possible. The student choosing this project should have basic experience of image processing and user interface design. Reference: [1] Yin Li, Jian Sun, Chi-Keung Tang, Heung-Yeung Shum. Lazy Snapping. ACM Transactions on Graphics (TOG) special issue: Proceedings of ACM SIGGRAPH, vol. 23, no. 3, July 2004, pages 303-308. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP3 Contact: Prof. Wang (wenping@cs.hku.hk) Mr. Dongming Yan (dmyan@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)
Multiscale Texture Synthesis	Example-based texture synthesis algorithms have gained widespread popularity for their ability to take a single input image and create a perceptually similar non-periodic texture of large scale. The aim of this project is to implement a multiscale texture synthesis algorithm. This algorithm takes as input a few low-resolution exemplar images at different scales, and creates infinitely detailed textures which are coherent and non-periodic. Some basic knowledge about C++ programming and image processing is required. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP7 References: [1]. LEFEBVRE S., AND HOPPE H. 2005. Parallel controllable texture synthesis. ACM SIGGRAPH, 777-786. [2]. Han C., Risser E., Ramamoorthi R., and Grinspun E. 2008. Multiscale Texture Synthesis. ACM SIGGRAPH. Contact: Prof. Wang (wenping@cs.hku.hk) Mr. Yufei Li (yfli@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)	Yu Hong
Vector Field Visualization	Vector field visualization is a hot research area in Computer Graphics, which can be widely used in the areas, such as physics, engineering, mathematics, etc, helping people to validate computational simulation. The aim of this project is to implement some popular texture-based and streamline-basedal algorithms of vector field visualization. The test on dynamic vector field data is expected. Some basic knowledge about C++ programming and image processing is required. Reference: [1]. S. Bachthaler and D. Weiskopf.Animation of Orthogonal Texture Patterns for Vector Field Visualization.IEEE Transactions on Visualization and Computer Graphics , pages 741-755, 2008. [2]. Abdelkrim Mebarki, Pierre Alliez, and Olivier Devillers. Farthest Point Seeding for Efficient Placement of Streamlines. In Proc. of IEEE Visualization, 2005. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP8 Contact: Prof. Wang (wenping@cs.hku.hk) Ms. Lin Lu (llu@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)
View-Dependent Model Simplification	Textured depth meshes and depth images are widely used to simplify drastically portions of the scene[1]. The simplification process is based on user view related parameters, which can provide nearly the same rendering results with less rendering time. This project aims on creation textured depth mesh for complex scene model. Reference: [1] Stefan Jeschke, Michael Wimmer,Textured depth meshes for real-time rendering of arbitrary scenes, Proceedings of the 13th Eurographics workshop on Rendering, 181 - 190. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP6 Contact: Prof. Wang (wenping@cs.hku.hk) Mr. Li Cao (lcao@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	1	Prof  Wang W (Between June 17-June 24, office hour or by appointment via email)
Visualizing Water Quality Variation using Google Earth	This project is aimed to provide high quality visualization of 3D water quality data around Hong Kong by integrating GIS data and 3D data visualization. Google Earth, which provides worldwide satellite images and terrain data, is an excellent source of GIS data. The project will develop a system on top of the Google Earth system to visualize simulated 3D water quality data. More details at the link: http://i2.cs.hku.hk/~gfxgroup/fyp2009/index.htm#FYP2 Contact: Prof. Wang (wenping@cs.hku.hk) Dr. Bin Chan (bchan@cs.hku.hk)	Attend Graphics course (0271) in the first semester of 2009-2010	2	Prof  Wang W (Between June 17-June 24, office hour or by appointment or email)
3D Tracking using Wii Remote	Wii is a video game console developed by Nintendo. One distinguishing feature of Wii is its wireless controller called the Wii Remote. The Wii Remote can be used as a handheld pointing device and detect movement in three dimensions. This allows players to experience a completely new interaction with the video games. In the project, we will first implement a 3D tracking system using the Wii Remote connected to a PC via bluetooth. We will then develop a user-friendly hand-guesture-based interface using the 3D tracking system.	C++ or Java programming skills	2	Dr  Wong Kenneth (make an appointment via email first)	Leung Wai Chun Hui Ying Kit (ykhui)
Building a multi-touch interface using a webcam	Traditional touch pads and touch screens can only respond to a single touch point. They are commonly used to replace the mouse as a pointing device in notebook computers and other user terminals (e.g. ATM). In recent years, multi-touch interface is becoming more and more popular (e.g. in iphone, macbook). A multi-touch interface allows multiple points of intersection, and can provide a more intuitive user interaction (e.g. resizing a photo by touching its diagonal corners and moving the touch points apart). In this project, we will construct a simple multi-touch interface using infra-red leds and a webcam, and implement an interface software for computing the coordinates of the touch points. We will also develop a user-friendly software for managing and editing photos using the muti-touch interface.	C++ or Java programming skills	2	Dr  Wong Kenneth (make an appointment via email first)	Chow Ho King Leung Fu Tuen
An Emulation Framework for Peer-to-Peer Applications	Large-scale peer-to-peer(P2P) applications have been successfully deployed over the Internet today. Based on the intriguing paradigm of using regular users' bandwidth to alleviate server load, new P2P applications are emerging quickly over time. To rapidly evaluate new P2P designs, it is desirable to implement them first in a controlled emulation platform that emulates real-world networking scenarios, before deploying them onto the real Internet. Due to the diversity of different P2P protocols, it would be an interesting and challenging task to develop an platform where various P2P protocols can be easily emulated. In this project, we aim to design and implement such an emulation framework for efficient implementation and evaluation of new P2P application designs. After the implementation of the basic emulation framework, we are going to show its usage with the prototype implemention of a modern P2P streaming application.	Multithreaded network programing	3	Dr  Wu C (5-7pm Tue June 16; 5-7pm Fri June 19)	Chan Ka Yui Chung Wai Yin (wyjchung) Wong Chi Long
Network Crawler for Peer-to-Peer and Social Network Applications	Various Peer-to-Peer (P2P) and social network applications have emerged over the Internet today. These new-generation applications have attracted millions of users around the globe, brought about large volumes of new types of traffic, and largely changed the traditional traffic pattern in the Internet. In order to understand their impact and to motivate the design of next-generation networks and applications, it is an important step to obtain a good understanding of the behavior and traffic pattern of the popular P2P and social network applications. Since most of the applications are implemented using proprietary software/code, the current technique to learn of their behavior is to implement "crawlers" which reach to users participating in an application, collect data from them, and use these data for behavior analysis. In this project, we aim to design and implement such a crawler program, which provides the framework to support effective crawling of representative P2P and social network applications. Such a crawler implementation would typically include reverse-enginneering analysis of popular P2P/social network applications, protocol messaging implementation, and then actual probing of the application users following their specific protocols. We will be targeting at popular applications such as Flashget, Xunlei or YouTube for our crawler implementation.	Multithreaded network programming	2	Dr  Wu C (5-7pm Tue June 16; 5-7pm Fri June 19)
ASAP: Availability System Abstraction Project			4	Dr  Yip Beta (by appointment)	Leung Wai Kin
A study of lightweight encryption scheme for mobile devices			1	Dr  Yiu S M (-)	Cheung Ka Ho Yim Lok Hin Fung Ying Ho (yhfung)
Algorithms for structural alignments	Given two strings (e.g. ABACCD and ABBCD), to evaluate how similar these two strings are, we can align them as follows: 1 2 3 4 5 6 A B A C C D A B B C _ D where column 3 can be treated as a mismatch, column 5 can be regarded as an insertion of space, other columns are matches. By giving a score for each matched column and penalties to both mismatches and insertions of spaces, we can quantify how good the alignment is. So, by finding the alignment with the highest score, we can tell how similar the two strings are. The project is to study this kind of alignment problems with additional constraints which are found to be useful in biological applications. Implenentation is required to evaluate the proposed solutions. The project can be done by 1 or 2 students, each student will be given a different variation of the problem to study.	Must be good at algorithms and programming	2	Dr  Yiu S M (Make an appointment (11 - 19 Jun) with me by emails)	Cheung Wing Yan (wycheung)
Implementation of Indexing data structures (for biological applications)	Searching patterns is a common operation in many applications. To speed up this operation, the data (e.g. a long string) to be searched will first be indexed. In this project, we study and implement some of the indexing data structures which are found to be useful in biological applications. No prior knowledge in biology is required.	Must be good at algorithms and programming	2	Dr  Yiu S M (Make an appointment (11 - 19 Jun) with me by emails)
Prototype for protocols in secure vehicular network	A vehicular ad-hoc network (VANET) is a mobile ad-hoc network for vehicles on roads. It is supposed that road side units (computers) are installed along the road and there is an onbroad unit (also a small computer) on each vehicle. Vehicles on roads and road side units can form a network and communicate. The purpose of such a network is to allow vehicles and road side units to communicate with one another so as to enhance driving safety. Messages can be broadcasted from a vehicle to nearby vehicles (and road side units) warning them about traffic accidents and road conditions. There are existing secure and privacy preserving protocols proposed for this kind of network. Roughly speaking, secure and privacy preseerving mean that messages sent by vehicles are authenticated and at the same time, routes of vehicles are difficult to track (in other words, the privacy of the drivers are protected). The projects is to implement and evaluate some of these solutions. This is a potential joint project with an IT company.	Must be good at programming and interested in security & cryptography	2	Dr  Yiu S M (Make an appointment (11 - 19 Jun) with me by emails)	Cheung Chung Wai (cwcheung) Ho Lai Wing Shum Ka Yiu Tang Man Hung (mhtang)
Prototype for protocols in secure vehicular network (Group 2)	A vehicular ad-hoc network (VANET) is a mobile ad-hoc network for vehicles on roads. It is supposed that road side units (computers) are installed along the road and there is an onbroad unit (also a small computer) on each vehicle. Vehicles on roads and road side units can form a network and communicate. The purpose of such a network is to allow vehicles and road side units to communicate with one another so as to enhance driving safety. Messages can be broadcasted from a vehicle to nearby vehicles (and road side units) warning them about traffic accidents and road conditions. There are existing secure and privacy preserving protocols proposed for this kind of network. Roughly speaking, secure and privacy preseerving mean that messages sent by vehicles are authenticated and at the same time, routes of vehicles are difficult to track (in other words, the privacy of the drivers are protected). The projects is to implement and evaluate some of these solutions. This is a potential joint project with an IT company.	Must be good at programming and interested in security & cryptography	2	Dr  Yiu S M (Make an appointment (11 - 19 Jun) with me by emails)
Reading confidential emails with mobile devices	This is a joint project with a local IT company. The objective of the project is to implement a feasible solution on mobile devices (e.g. black berry, mobile phone) to read and send confidential (encrypted) emails. The difficulty of the project lies on the limited computational power of the mobile devices which make it difficult to perform encryption/decryption operations. Interested students are recommended to take CSIS0327 (Computer and Network Security).	Must be good at programming, interested in security & cryptography	2	Dr  Yiu S M (Please make an appointment (11 - 19 Jun) with me by email )	Ng Ho Lam Tsang Chi Fung
Searching encrypted data			1	Dr  Yiu S M (-)	Cheung Wai Lok Choi Chak Yung (cychoi) Lau Kim Yuk