[University of New England] [Turing]

Charles Watson School of Science and Technology The University of New England Armidale NSW Australia

---

Office:	MCS, Room 155
Phone:	+61 2 6773 2288
Fax:	+61 2 6773 3312
Email:	cwatson7 @ une.edu.au

Research Interests

· Computational Genomics

· Algorithmic Complexity

· Cryptography

· Information Security

· Signal Processing

· VLSI Design

· Automated Trading

· Online Learning

Ongoing Research Projects

1. To develop efficient methods for haplotype mapping of large pedigrees with high linkage. The feasibility of using large numbers of genetic markers (>100,000) to track and evaluate genome segments for their impact on commercially important traits is now becoming evident.  This is particularly true in the dairy industry where accurate evaluations on young bulls obviates the need to wait till they are grandfathers before getting milking data on their daughters.  This project aims to reduce costs significantly by reducing the numbers of markers required to perhaps 5,000.  Pedigree information gives the prospect of tracking the fate of segments of DNA as they pass down the pedigree, identifying where the recombinations take place that shuffle the genetic material.  This means that much larger DNA segments can be tracked, requiring fewer markers.  The methods to be developed will infer haplotype states (linkage phases of markers) in foundation animals (those with no known parents), and deduce most probably pattern of recombination and transmission of genetic material down the known pedigree. An object oriented model is run on a multithreaded architecture. This Agent model as an abstract object (class) with internal estimates of haplotype likelihoods and links to parents and offspring. Each individual is represented by an instance of this class. Constraints, such as homozygous parent SNPs, are propagated to related (linked) objects by message passing. The state of haplotype likelihoods in each individual is analogous to a hidden Markov model converging to a stable state by relaxation.

2. To investigate emerging Bioinformatic techniques and their application to practical problems in genetics, applying experience in software development, signal processing and cryptanalysis to the interface niche between the disciplines of Computing, Mathematics, Statistics and Genetics.

3. To develop my software implementation of conditional entropy indexing to use self-index compression for optimal retrieval speed and memory usage. This will overcome the excessive memory use of the current implementation enabling the systematic comparison of genomic information content on a larger scale than current practice in genomics. Recent developments in self-index algorithms (Navarro and Makinen 2007) provide memory use similar to the best existing compression techniques. A quad-core, XEON with 8Gbyte of RAM and 1.5 TeraBytes of disk storage was purchased and installed. Various software including Pizza-Chilli API of indexes, wwwBlast was installed. The indexing software and test databases developed in 2007 were also installed and benchmarked. This implementation of a self-index server for fast access & efficient memory utilises recent algorithm development to demonstrate large scale web/data services. This timely project should lead to commercial licensing with a high expected return on investment.

4. To develop a self-index to improve memory usage of the "Killer-Application for Plagiarism Detection". The expected outcome is a fully developed, tested and documented implementation of the Conditional Entropy algorithm with a compressed self-index to provide optimal retrieval speed and memory usage. This program will be made available as web service for high throughput workflows. The algorithm will be published as "A conditional entropy web service for high throughput pattern analysis". This software will be used as a plagiarism detection filter for electronic assignment and thesis submission. The collective use by many institutions will help maintain best scholarly practice and identify sources of plagiarism proliferation.

Recent Research

Web Services: Content and bandwidth on the Internet is increasing exponentially, creating the need for automated management of distributed information. Interoperability between information systems is facilitated by open standards, such as Universal Description, Discovery and Integration (UDDI), Simple Object Access Protocol (SOAP) and Web Services Description Language (WSDL). Document-type definitions (DTDs) and XML Schemas enable autonomous services to exchange information. I am interested in testing and deploying new services to meet future requirements of e-Commerce using WebSphere and J2EE servers. I developed a distributed computing test-bed for running secure web services to wired and wireless clients using various security protocols. The servers include IIS, WebSphere, Tomcat, LDAP, and J2EE, and the wireless technologies include J2ME, IrDA, BlueTooth and 802.11b/g. The security technology (smart-cards, device authentication, WS-Security, SAML, Public Key Infrastructure and packet filtering) allows confidential, reliable and authenticated communications between distributed mobile participants. I recently integrated a Rule Based Access Control system with Collaxa BPEL for modelling e-business workflows.

Java Multiprocessor: This project draws on previous research in VLSI processor design, and a hierarchical cached memory system I designed. It combines large, low-cost shared memory with small fast caches to provide a cost effective, high performance memory system. This architecture has been simulated for the declarative languages LISP and PROLOG. This architecture now applied to Java. The processing elements (PEs) execute the Java Virtual Machine (JVM), and light-weight threads are implemented by extending the process stack from shared memory to local cache on the PEs, where threads are executed.

Web Ontology Generation: Algorithms for generating knowledge structure or semantics from unstructured data will transform the Internet. Agent-like search engines deliver documents, and extract knowledge. Computers are shifting from single isolated devices to entry points in a worldwide network of information and business transactions. I installed WebSphere Application Server, the WS SDK (with eclipse) DB2, J2EE, IIS, J2ME, Ontomat, RDF-crawler, Koan, Jena, bouncycastle. These applications are integrated into a platform for distributed, intelligent data-mining.

Secure objects: The aim is to harden software objects in hostile environments. These hardened objects can only be accessed/activated via an authorized key. Keys are generated in sequence by a key generating authority corresponding to each object instance. The keys provide zero knowledge of the encryption or the content of the objects. Sequential activation gives an account (audit trail) of usage that can be logged by the key generating authority. Key generation can also be delegated so a lease for a specific number of accesses can be given to a secure host. The security mechanism is hierarchical so there can be objects within secure objects. The delegated authority for nested objects can be locked within the parent object. These protected objects can safely remain dormant in a hostile environment or be transmitted over public communications infrastructure without the risk of tampering. This process can also be applied to digital rights management. I'm interested in exploring these ideas further and demonstrating an implementation using WS-Security and Java CardLets.

Cryptanalysis: This work involves a new algorithm for predicting the output of ciphers. A computer model of the cipher is initialized to the zero knowledge state.  For each iteration, known bits of state are set and after a round of execution the difference between the predicted and expected output state is back-propagated to the inner state of the model. The output either converges to an error-free predictor or fluctuates near the zero knowledge state. Simulated annealing can be used to assist convergence. By using a finite state model, this algorithm can be applied to any black-box cipher, where the inner structure is unknown.

Automated trading: The development and testing of algorithms to exploit inefficiency in financial market, including a set of basis functions formed by the difference of exponentially weighted means. The exponentially weighted variance is used to continuously measure market volatility and avoids the "ballooning" of Bollinger bands at the start of a trend. These functions scale with time to embrace the fractal self-similarity of markets which tend to trend retrace. The algorithm signals the start and finish of trends and executes faster more accurately than a human traders, thereby providing market liquidity, minimal volatility and market efficiency while extracting surplus equity.

Recent Publications

Refereed Journal Articles

"Intron Phase Patterns in Genes: Preservation and Evolutionary Changes", The Open Evolution Journal, Vol. 1, pp.1-14, 2007. Authors: A. Ruvinsky, C. Watson http://www.bentham.org/open/toevolj/openaccess2.htm

Refereed Conference Papers

"Creating a User-steerable Media Presentation-system as a Canadian/Australian Distance-Learning Research Project", Association for Canadian Studies in Australia and New Zealand, July 2010.

"Anomalies in multidimensional contexts", Data Mining 2009.

"Conditional Entropy for DNA Analysis", International Symposium on Information Theory and its Applications, Dec 2008, Auckland, New Zealand.

"Differential Convolution for Medical Diagnosis", Proceedings 5th International Conference on Information Technology and Applications, June 2008.

"Wireless location aware applications for mobile commerce", Proceedings of the Fifth Asia Pacific Industrial Engineering and Management Systems Conference, Vol.5, pp274-290, Dec 2004, QUT.

Seminars and Presentations

"An algorithm to exploit market inefficiency", Friday 14th November 2008. http://mcs.une.edu.au/~cwatson7/I/AutomatedTrading.pdf

"Clever Networks", Friday 29th June 2007. http://mcs.une.edu.au/~cwatson7/I/CleverNetworks.ppt

"Virtual Infrastructure", Friday, 25th May 2007. http://mcs.une.edu.au/~cwatson7/I/VirtualInfrastructure.ppt

"Gene mutation and evolution of the kinectin coding gene, KTN1", AGBU Seminar on Tuesday 15th, May 2007. http://mcs.une.edu.au/~cwatson7/I/KinectinPhylogeny.htm

"Statistical forecasting for predicting markets, weather, speech and other phenomena", Friday, 30th March 2007. http://mcs.une.edu.au/~cwatson7/I/TimeSeriesPrediction.ppt

"An efficient algorithm for analysing large data sets, by example", Friday, 23rd February 2007.  http://mcs.une.edu.au/~cwatson7/I/ConditionalEntropy.html

"Platonic Solids", UNE Open Day, September, 2006. http://mcs.une.edu.au/~cwatson7/PlatonicSolids.htm

"Zero Knowledge Cryptanalysis", ISRC Seminar, QUT, Brisbane, May 2004. http://mcs.une.edu.au/~cwatson7/Seminar/ZKC.ppt

Earlier publications...

"The Computer Analysis of Polyphonic Music", Proc. 22nd ANZAAS Conference, Brisbane, May 1981.

"Algorithms for the separation of Superimposed Tones", Proc. lst Conference on Music and Technology, Melbourne, August 1981.

"SPRINT Functional Specification", Internal working document, VLSI-IWD-82-12-1, CSIRO. "AUSMPC Project Report", May 1982, CSIRO VLSI Program.

"A Floorplan Algebra and Composition Paradigm for VLSI Design", Technical Report, VLSI-TR-83-3-1, CSIRO.

"A Proposal for a Silicon Implementation Language (SIL) Technical Report, VLSI-TR-833-25 CSIRO.

"An Algorithm for the Correct Interconnection of VLSI-Level Macrocells", Proc. VLSI Pacific Asian Region Conference, Melbourne, May 1984.

"A Declarative Design Approach for Combining Macrocells by Directed Placement and Constructive Routing", Proc. 21st Design Automation Conference, New Mexico, June 1984.

"Speaker-Dependent Word Recognition - A Case Study of a Custom VLSI Chip Development", Proc. IEEE International Conference on Computer Design: VLSI in Computers, New York, Oct 1984.

"The Computer Analysis of Polyphonic Music", Ph.D. Thesis, University of Sydney, 1985.

"An Algorithm for the Correct Interconnection of VLSI-Level Macrocells", Journ. Electrical and Electronics Engineering, Australia. Sept 1985.

"VLSI Design and the 5th Generation Project in Japan", CSIRO Trip Report, 1986.

"The Design of a Parallel LISP/Prolog Machine", Internal Working Document, Div. Information Technology, CSIRO, 1986.

"VLSI Design", First ASEAN Science and Technology Conference, Kuala Lumpar, April 1986.

"A Custom IC Design for a Multi-Memory Mapping Unit", Proc. Aust. Microelectronics Conference, Adelaide, May 1986.

"Automated Layout of CMOS Circuits", Proc. Aust. Microelectronics Conference, Melbourne, April 1987.

"A Tool Set for Correct-by-Construction VLSI Design", Proc. Electronic Design Automation Conference, Wembley, England, July 1987.

"A CAD System for the Rapid Design of Custom Integrated Circuits", Proc. NELCON, Auckland, New Zealand, September 1987.

"A Rapid Design System for Custom Integrated Circuits", Digest of papers IREECON, Sydney, September 1987.

"A Neural Network Architecture", Proc. Aust. Symposium on Signal Processing and Applications, Adelaide, April 1989.

"Neural Accelerator Board (NAB) Specification", ITD Technical Report, June 1989.

"A High Performance Systolic System for Image Processing", Proc. Image Processing and the Impact of New Technologies, Canberra, December, 1989.

"A Custom VLSI Design for High Speed Emulation of Neural Networks", Proc. First Aust. Conference on Neural Networks, Sydney, January 1990.

"A Neural Accelerator". Proc. Aust. Microelectronics Conference. Adelaide. July 1990.

"A Systolic Neural Network Image Processing Architectutre". Proc. International Conference on Computer Systems and Software Engineering. The Netherlands. May 1992.

"Motion Analysis using the Neural Accelerator Board". Proc. International Conference on Pattern Recognition. The Netherlands. August 1992.

"Cryptographic Application of the Fundamental Finite Field". Published January 1999 at http://www.geocities.com/CapitolHill/Lobby/7077/zero.html

Teaching at UNE

COMP 100 Introductory Information Technology 2000-2012

Introduction to Business Information Technology (COMP100/500) is a one-semester unit offered in both first and second semester. It has been designed to provide essential knowledge for and give a firm foundation to students who expect to make substantial use of computers and computer-based information systems in their careers.

The unit involves instruction in the basic operation of personal computers and introduces students to word-processing, spreadsheets and databases. It does not assume any prior knowledge of computers. The unit aims to provide computer literacy skills that will prove essential for the completion of university courses and future work environment where increasing emphasis is being placed on information technology. This is online course develops computer skills including; participation in online discussions, electronic assignment submission and answering computer generated quizzes.

COMP 320 Computer Networks 2012

The aim of this unit is to study the foundations of communication networks in order to follow and contribute to their evolution.

Topics include:

• Application layer: the web, email, socket programming
• Transport layer
• Network layer
• Data link layer
• Physical layer
• Local area networks
• Wireless and mobile networks

COMP 321 Wireless Networks 2006-2008

Comp321 provides a hands-on guide to planning, designing, installing and configuring wireless LANs that prepares students for the Certified Wireless Network Administrator (CWNA) certification. The unit offers in-depth coverage of wireless networks with extensive step-by-step coverage of IEEE 802.11b/a/g/pre-n implementation, design, security, and troubleshooting. Material is reinforced with hands-on projects at the end of each chapter from two of the principal wireless LAN vendors, Cisco and Linksys.

Specific topic coverage includes:

• It’s a Wireless World
• Wireless LAN Devices and Standards
• How Wireless Works
• IEEE 802.11 Physical Layer Standards
• IEEE 802.11 Medium Access Control and Network Layer Standards
• Planning and Building a Wireless LAN
• Conducting a Site Survey
• Wireless LAN Security and Vulnerabilities
• Implementing Wireless LAN Security
• Managing a Wireless LAN
• Network Settings and Wireless LAN Troubleshooting
• Personal, Metropolitan, and Wide Area Wireless Networks

COMP 322 System and Network Administration 2006-2009

This unit teaches administration of computer networking hardware and software concepts including protocols, network design and implementation, and troubleshooting and support. Hands-On Projects, and Case Projects allow students to practice skills as they are learned, enabling them to administer networks in the home, small businesses, and large enterprises. Students use software from several vendors including Windows XP, Server 2003, UNIX-based and NetWare-Based administration software. On-The-Job stories each week provide a real-world context for the technology and concepts presented.

Learning Outcomes : Upon completion of this unit, students will be able to:

1. install, configure and administer WindowsXP,

2. plan and build a computer Network ,

3. implement Network Security,

4. administer networks in the home, small businesses, and large enterprises,

5. use Windows Server 2003, UNIX-based and NetWare-Based administration software,

6. Comp522 students will also be able to demonstrate application of this technology to their professional practice.

Teaching Methods and Assessment :

This unit has fully online teaching and assessment. Internal students will have addition face-to-face support with a 1 hour lecture and 2 hours of practical work each week. Weekly PowerPoint lectures are delivered over the Internet. Quizzes and Project work are submitted via the Internet. The quizzes are assessed automatically.

60% of the grade is based on quizzes. Twelve weekly electronic quizzes corresponding to each chapter give immediate feedback on completion. Each instance of the quiz has a different random selection of questions to discourage collusion. The quizzes will assess the student’s understanding of network administration.

40% of the grade is based on keeping an electronic project notebook with the results of Hands-on Projects and Case Projects. The final project report will be less than 5,000 words and will demonstrate the student’s skills of administering computer networks. Comp521 students will also demonstrate application of this technology to their professional practice.

COMP 411 Information Technology for E-Commerce 2000-2005

Comp411 was rewritten in 2004 with a new text-book to make the subject more relevant to business students and to bring the course up-to-date with emerging technologies. The new unit gives students hands-on experience in such areas as web services, Internet publishing, financial transaction security and online trading. The 3 new assignments are:

Assignment 1) Online trading - The aim of this assignment is to gain experience using online trading systems. Without using real money, students use a demo account that works the same as using a real $25,000. Students submit a Forex trading report for the month and a word document describing experiences, strategies and outcomes.

Assignment 2) Security/Payment Systems - to investigate use of SET and SSL, firewalls, security threats, including viruses, worms, or Trojan horses and use PGP software for secure communications.

Assignment 3) E-business Implementation - to write a development plan or system specification to describe the business process workflow for an enterprise or design and implement a demonstration prototype for the proposed e-business system.

COMP 516 Principles of Object-Orientated Programming Languages 2007

Comp516 covers advanced topics in object-oriented programming including type abstraction, subtyping, type-safety, container types, concurrency, components, persistence and security. The emphasis is not on a particular programming language but on underlying principles and alternatives in various programming languages and in the programming language literature.

COMP 290 Internet Security 2008-2009

This unit studies the principles and methods used for secure access and transfer of information on the Internet. Topics include:

Cryptography, Message Digests, Public Key Infrastructure, Authentication Systems and standards, E-Mail and Web Security.

FSC102 Forensic Computing 2008-2011

Forensic Computing is a module in the Bachelor of Criminology unit, FSC102. FSC102 describes the role of the forensic scientist and discusses the principles of crime scene investigations. The latest investigative methods and techniques for bloodstain pattern analysis, computer crime, firearms, drugs, toxicology and DNA fingerprinting are discussed. The Computer Forensics case study reports the forensic analysis of a computer system with sufficient detail for an expert to repeat the analysis and clearly present the results at the level of understanding of a jury of lay people. The September intensive school has lectures on Document and Voice recognition, Computer Forensics and the Internet. The school uses the 40 computers in MCL1 and 2 with Internet access and Office software and 4 forensic computers in G2 lab.

Research Links

Free On-line Dictionary of Computing

Computer Science Bibliography Collection

The MIT collection of links Computational Linguistics

Stanford laboratories

Computer Science, Research

KSL, the Knowledge Systems Laboratory

RPL, the Rapid Prototyping Laboratory

Computer Sciences Neural Nets Research Group UTexas

Laboratory for Artificial Neural Systems (LANS)

Association for Computing Machinery (ACM)

Electronic Commerce http://www.elib.org/

Handheld Computing

Computer-Mediated Communication (CMC) Information Sources

JavaScript Source hundreds of free JavaScripts

ABC News

US National Information Infrastructure

The World-Wide Web Consortium

NASA

The National Security Agency Web Page

Firewalls FAQ

Computer Security FAQs

Pretty Good Privacy [PGP] FAQs

The World Wide Web Security FAQ

Cryptography FAQs

Computer Virus FAQs

The Career Information Clearinghouse

Yahoo! Australia & NZ - with a uniquely Australian and New Zealand perspective

Google Computer list

European language translation with Systran® http://babelfish.altavista.com/translate.dyn

Verisign/RSA

Don't mess with my circles

Archimedes was killed by a Roman soldier during the sack of Syracuse during the Second Punic War, despite orders from the Roman general Marcellus that he was not to be harmed. The Greeks said that he was killed while drawing an equation in the sand; engrossed in his diagram and impatient with being interrupted, he is said to have muttered his famous last words before being slain by an enraged Roman soldier: Μη μου τους κύκλους τάραττε ("Don't mess with my circles"). The phrase is often given in Latin as "Noli turbare circulos meos".

Archimedean Solids

Platonic Solids

Information Research Group

http://mcs.une.edu.au/~cwatson7/I/

http://blog.une.edu.au/informationresearchgroup/

The objectives of the group are to increase impact and visibility of the disciplines of Mathematics, Statistics and Computer Science within the university, increase publications, to offer more frequent seminars, invite and involve all staff and students at UNE and the broader community to the seminars, to increase HDR enrolments and completions, and to obtain grants wherever possible to benefit both SMSCS and UNE.

• It was agreed that the main task is to identify strengths leading to commercial projects.
• It was decided to meet on the last Friday of every month.
• It was decided that we remain an informal group until objectives were defined and projects were identified and developed. At that point a formal approach to the university will be considered.

Charles Richard WATSON, Ph.D., B.Sc.(Hons.1st) School of Science and Technology The University of New England Armidale, NSW 2351.

---

Created and maintained by Charles Watson. University of New England, Armidale, NSW 2351
URL: http://mcs.une.edu.au/~cwatson7 Last revised: 21^st May 2012