My supervisor was Professor Bernard Pagurek
I am a second degree black belt in karate (May 2011).
Artificial Intelligence, Social Computing, Information Systems, Distributed Computing, Internet Applications and Web Technologies, Machine Learning, Mobile Agents, Peer-to-Peer Computing, Swarm Intelligence, Evolutionary Computing
Advisory Board Memberships
Embotics: September, 2007 – continues
Immortix Corporation: February, 2002 – September, 2002
Texar Corporation: September, 2001 – June, 2002
March, 2006-October, 2006: Embotics Corporation, Ottawa, Canada
Embotics is the software-only company derived from the Symbium Corporation. I acted in the capacity of Chief Technology Officer for the corporation. I left at the end of my leave of absence from Carleton. I am currently an advisory board member.
January, 2004-March, 2006: Symbium Corporation, Ottawa, Canada
As Chief Technology Officer, I was responsible for intellectual property portfolio management, technology trend assessment, strategic technology selection and product roadmap.
September, 2001-continues: School of Computer Science, Carleton University, Ottawa, Canada
I am a Professor of Computer Science, primarily teaching computer languages, distributed computing and web technologies. My research interests include network science, evolutionary computing and autonomic computing. I am currently actively undertaking research in: Trust and reputation systems, influence measurement in social networks and its role in the control of agent-based societies. See publications page for details.
A number of patents resulted from this research.
July, 1999-June 2002: Texar Corporation, Ottawa, Canada
As Chief Scientist, I was responsible for R&D. I built, and managed, a world-class team of more than 30 developers. I was responsible for technology assessment and selection and tracked emerging technologies in the areas of information security, peer-to-peer computing and software agents. I managed external research projects at Carleton University in the areas of Distributed Search, Peer-to-Peer computing, Distributed Functional Programming and next generation application execution environments involving software agents. I actively participated in the development of SecureRealms and peer-to-peer technologies and supported sales activities as an evangelist for Texar. I also advised Texar on issues pertaining to information security and peer-to-peer computing during my tenure on the advisory board. I was a member of the P2P Working Group subcommittee on security and consulted with William Yeager (author of the Poblano trust architecture) on authorization issues for JXTA.
A patent resulted from the work on SecureRealms.
August, 1997-July 1999: Systems and Computer Engineering Department, Carleton University, Ottawa, Canada
I was a Senior Researcher within the Perpetuum Mobile Procura group, a group that investigates the application of mobile agents to problems in the area of network management. I participated in student supervision, reviewed group paper and report output, and provided technical program input regarding aspects of the program's research direction. I lectured on mobile agents, aspects of network management and data structures when asked to do so. I was the technical prime on the architectural evolution of the mobility toolkit and provided support for the existing implementation. I formally co-supervised four graduate students in the areas of: mobile agents for IP telephony advanced services, plug-and-play networks and hot swapping technologies for next generation applications.
A patent resulted from the IP telephony work.
My principal role was in research of mobile agent technology; specifically, how mobile agent activities can be coordinated by exploitation of the behaviour of societies of simple agents. In the systems that I considered, problem solving is considered an emergent property of the system; i.e. the interactions of many agents with their environment solve the problem but no one agent has an overall view of the system or problem to be solved. Local, rather than global, interactions determine the behaviour of the system.
Upon completing my Ph.D., I was appointed as an adjunct professor within the department.
November, 1986-April, 1999: Nortel Technology, Ottawa, Canada
January, 1995-December, 1996
I was on secondment to the Harlow U.K. laboratory, working within the Software and System Engineering (S&SE) Department. Approximately 50% of the secondment was devoted to the exploration of the use of Genetic Algorithms (GAs) in problems in Telecommunications. This included the generation of several prototypical applications for ring design and community of interest visualization (essentially a clustering problem). A number of papers were written concerning this work. I was also involved in the continued development of RouteFinder - a router that uses GAs in order to provide balanced routing solutions in networks and have embedded all of these applications in a Smalltalk framework for demonstration purposes. An agent-based solution of a routing problem in SONET/SDH networks was also built.
The remaining 50% of my time was spent working on agent-based Alarm Correlation. The result of this work was the design and implementation of an Alarm Correlation engine used for the Concorde switch product. This work was performed exclusively in Smalltalk and involved embedding a production rule environment seamlessly into Smalltalk (based upon NeOpus). A rule compiler was written a part of this activity.
Several patents resulted from this work.
July, 1993-December, 1994
Member of the Computer Research Laboratory (CRL). I worked on a configuration problem relating to network design and built a costing and provisioning system for ATM networks. I also built a Smalltalk system for network design called DesignMate which was useful in my work at the Harlow laboratory.
August, 1992-July, 1993
I worked as part of the Data Packet Network Network Management (DPN NMS) group. I wrote a proposal for an engineering design and analysis system for DPN-100 networks and provided consulting for the Expert Advisor. I also added an optimization feature for a routing package used in engineering DPN-100 networks. A genetic algorithm solution to the same optimization problem was proposed but not implemented.
August, 1991-August, 1992
On sabbatical at Carleton undertaking research in the area of Artificial Life and Evolutionary Computation. I worked with Professor F. Oppacher during this time.
November, 1988-August, 1991
Principal Architect for the Expert Advisor. The Expert Advisor was an alarm correlation engine used for diagnosing problems in a DPN-100 network. The expert system was large, consisting of greater than 18,000 rules and problem structures. Considerable work went into a number of prototypes in order to arrive at the product knowledge representation. A novel solution to the common problem of complex rule interaction was designed and implemented through a use of message passing techniques. The system was deployed in a large number of sites during the period 1992-1995. Several papers have been published describing this system.
A full customization environment was implemented for this system in order to allow end users to modify the delivered expert system and prove the changes in a simulated environment before integration into the operational network management system.
This system was the subject of a Canadian A.I. magazine article on Canadian A.I. success stories written by Dr. Suhayya Abu-Hakima formally of the NRC. A number of papers relating to this work can be found on my publications page.
November, 1986-November, 1988
I worked on DMSpulse - an application which charted performance measurements from a DMS 100. This application was written in C on a PC. I also worked on SysPlan - a configuration engine built in Prolog on a PC. This system was used to configure software systems.
I managed the Modelling Research Group, being responsible for the generation of PC-based software for the modelling of IBM mainframe performance. This position required a knowledge of Queuing Theory, Probability and Statistics. I managed a group of two other Capacity Planning staff. As a major part of my work, I was responsible for the research into algorithms that could be used to model a complex mainframe computer system accurately while still being able to run on a PC. Prior to my departure, work began on an expert system product to be used in the tuning of the IBM MVS operating system. This expert system prototyping began using the TIMM expert system shell.
During my year as a consultant, I developed signal processing software (based upon alpha-beta filters) for a gun fire control system. I also developed a package which was used in fitting curves to experimentally-derived shell trajectory data. This software was written in Pascal on a VAX running VMS.
I wrote small software packages, in Pascal, for various John Bell clients. The biggest project was for a defence contractor where a trials data base was designed.
As part of the Systems Analysis Group, I built a model to process telemetry data from a missile, filter it and use it to verify a 6 degree-of-freedom model of the missile. I also designed novel algorithms for a gun fire control system and investigated missile control algorithms for the next generation of nuclear missile deterrent.