Organized by:

Tutorial 1
Introduction to ANNs, data preparation techniques & application development
  Waseem Asrar Ahmed
NED University of Engineering & Technology, Pakistan
Waseem Asrar Ahmed is a Computer Systems Engineer and an adjunct faculty member at Hamdard University & NED University of Engineering & Technology, Karachi. He conducts courses of Artificial Intelligence, Intelligent Systems, Neural Network Computing & Real-time Embedded Systems, etc. He has above 11 years of experience that includes 7+ years field experience in developing real-time data acquisition systems for industries. Also serving as Consultant Engineer for developing AI systems, he has contributed several articles & research papers and participated in many international events on the subject. He has conducted several workshops on AI, NNC & Fuzzy applications.

The Tutorial focuses on data preprocessing for effective training of neural nets. Selection of input vectors and preprocessing of data is followed by the training assignment.
Participants may preferably have an idea of basic functionality of neural networks.

      a) To introduce the idea of using operations of our brain as a model for computing.
      b) Awareness of role which Artificial Neural Networks can play in problem-solving.
      c) Discuss a few types of ANNs & their applications.
      d) Identify & judge where ANNs are useful in problem-solving.
      e) Design, construct and training of ANNs.
      f ) Data preprocessing for an effective training of ANNs.

Neuron modeling    
Learning algorithms illustrations    
Innovative real-world applications    
Engineering & business applications    
Case studies and demonstrations    
Hands-on sessions on design, construction & training of neural nets    
Data-preprocessing techniques for neural nets    
Intelligent preprocessing techniques & tools, etc.    
Target participants
IT Professionals including Systems Analysts, Programmers, Systems Engineers, Project Managers, faculty members & students of IT institutes.

  Tutorial 2
MEMS enabled Microsystems: Cogent sensing and intelligent applications
Tutorial delivered by:
Dr. Elena Gaura
Coventry University, U.K.
Co-authored by:
Dr. Robert Newman
Coventry University, U.K.
Elena Gaura is a Senior Lecturer with the School of Mathematical and Information Sciences, Coventry University, specialising in MEMS based intelligent sensors, computer hardware, Artificial Intelligence and pervasive computing. She is the leader of the Informatics, Media and Design (IMD) research group.
She is a member of the UK Engineering and Physical Sciences Research Council College of Peer Reviewers and a member of the Program Committee for Nano Science and Technology Institute’s NanoTechnology Conference and Trade Show.
Presently, her research interests pursue the issues of MEMS sensor systems design (to include microsensors-Artificial Intelligence integration, sensor fault detection, self-diagnosis, microsensor applications to safety critical and biomedical systems and sensor networks. The work explores the new avenues brought about by MEMS technology to enhance the functionality of micro measurement systems, develop new techniques for integrating sensors, actuators and control functions and ultimately aim at designing autonomous systems which can sense, think and react to their working environment. Much research effort is currently focused on theoretical and practical design aspects for very large networks of autonomous MEMS based sensors.
Dr. Gaura graduated with a MSc in Applied Electronics from Technical University of Cluj, Romania, in 1991. Her research interests were in the field of Artificial Neural Networks (ANNs), with a focus on ANN VLSI implementations. She worked for Brunel University, Uxbridge, UK and subsequently for Rutherford Appleton Laboratory. In 1998 she was offered a HEFCE PhD grant at Coventry University. She was awarded her Ph.D. at Coventry University, School of Engineering, in April 2000, with a thesis entitled: ‘Neural Network Techniques for the Control and Identification of Acceleration Sensors’. She has been invited to Chair ANNs and Intelligent Control related sections at various conferences and is a member of the Program Committee of the International Conference on Artificial Neural Networks and Genetic Algorithms (ICANNGA). She has received several IEEE, Royal Society and Royal Academy of Engineering grants for presenting her research work at various conferences and academic institutions. She is the organiser of a Special Track on “Smart MEMS and Sensor Systems” at the largest Nanotechnology American Conference, Nanotech. She is a Journal reviewer for the Journal of NeuroComputing (Elsevier Science), the Mechatronics Journal (Pergamon Press, Oxford, U.K.) and IEEE Transactions on Control Systems Technology.

Robert Newman is currently Head of Computer Science at Coventry University. He has managed and produced successful research proposals of a value greater than £500 000 from the EPSRC and the European Union and has included the support of a number of major industrial concerns. He holds a BSc in Physics from Birmingham University and a PhD in Computer Science, in the area of safety critical systems, from Coventry University. He has produced 50 refereed publications and has served on the programme committee of five major international conferences and in 1991 was awarded a patent for the design of an autonomous intelligent sensor, one of the earliest in this field. The major theme of his research is pervasive computing, particularly the system design of distributed intelligent systems, and their application and the use of formal methods and systems science in their design and specification. His research has involved major collaborations with the European aerospace and automotive industries over a period of ten years, and has included four major projects, three funded by the European Union and one by the EPSRC, with industrial collaborators including Ford, Rolls-Royce PLC, Volkswagen, BMW, BAE Systems and EADS. He is also a member of the UK Department of Trade and Industry Foresight Vehicle Steering Committee for Design and Manufacturing Processes (DMAP).

To present the directions of research, development and technological evolution for Electro Mechanical Microsystems, and in particular microsensors. The development of MEMS devices has generally followed a bottom up methodology, reaching now a stage where the capabilities of the devices could be used much more effectively in systems designed from the top down to include them. A holistic view of the requirements of MEMS based systems and the capabilities of the microdevices must be taken if such systems are to deliver the promise that was expected. This tutorial provides the integrative perspective required for workers in all areas of the field, to enable them to appreciate the system level design issues leading to breakthrough applications, particularly in the area of large sensor networks.

The tutorial would be of interest to:

Control/system engineers and robotics specialists who use sensors as part of process, plant or robot control. The tutorial will inform on the state-of-the-art in sensing and the possibilities opened by advances in Microsystems design.
Specialists in Artificial Intelligence who will find great openings for AI applications in future cogent sensing systems. AI is likely to play an important role in the information extraction/management and the realization of large scale networks of sensors.
Circuit designers whose work is in the areas of electronic interfacing of MEMS, calibration, electronic design for performance enhancement, robustness and reliability. The tutorial will be of interest from the viewpoint of system partitioning and hardware design of intelligent nodes, node design for dedicate collaborative problem solving.
MEMS technologists/designers/developers as they need to have an awareness of the design constrains of the systems which will use their devices. Such concerns are likely to influence the specification and detail design of the microdevices and the processes used to fabricate and package them.
Specialists working at system level in sensor networks. This tutorial will allow them to understand how their specialism relates to the application and device level specialisms. Allocations of functions at different levels in the overall needs to be considered at this level to allow optimum design.
Application developers considering use of networks of intelligent MEMS devices who will need to understand and be able to handle the complexities of design of such systems.
Click here for more details
Copyright © 2003 INSTICC