Internet Technology in Laboratory Modules

for Distance-Learning

 

A project financed by a contract with Nordunet2

 


Participants and Organization

Principal Investigator: Tor A. Fjeldly, Professor of Electronics
UniK - Center for Technology at Kjeller
Norwegian University of Science and Technology (NTNU)
N-2027 Kjeller, Norway
Tel: +47 64 84 47 27 , Fax: +47 63 81 81 46
E-Mail: torfj@unik.no, URL: http://www.unik.no/~torfj/

 

Co-Principal Investigator: Kjell Jeppson, Professor of Electronics

Chalmers University of Technology

Dept. of Microelectronics ED

Solid-State Electronics Laboratory

S-412 96 Göteborg, Sweden

Tel: +46-31 772 1856,  Fax: +46-31 772 3622

E-Mail: kjellj@ic.chalmers.se, URL: http://www.ic.chalmers.se/~kjellj/

 

Additional participants: Graduate and undergraduate students at the Norwegian University of  Science and Technology, Center for Technology at Kjeller, Norway and Chalmers University of Technology, Göteborg, Sweden.

 

Collaboration:

Prof. Michael Shur, Rensselaer Polytechnic Institute, Troy, NY, URL: http://nina.ecse.rpi.edu/shur/.

Additional university groups.

 

 

Administrative contact:
Mr. Ivar Jardar Aasen
Director, UniK - Center for Technology at Kjeller
N-2027 Kjeller, Norway
Tel: +47 64 84 47 05 , Fax: +47 63 81 81 46
E-Mail: ivara@www.unik.no

 


Summary

The objective of this project is to establish user-friendly and efficient technology for interactive, on-line operation of remote education laboratory equipment and experiments, utilizing the Internet and the World Wide Web. We have already demonstrated the suitability of this technology in preliminary versions of the remote lab concept, developed in collaboration between the NTNU and Rensselaer Polytechnic Institute (RPI) in Troy, NY. So far, the remote lab has been dedicated to semiconductor device characterization. In the last year, it was tested in distance-education courses on semiconductor device technology with good results. Also, the test revealed how the system could be improved in several ways to give a more genuine lab experience. Central issues within the proposed project are to increase the flexibility in configuring the experiments from the client side, to improve the user interface, and to increase the number of experiments.  Eventually, based on this concept, lab courses and course modules within many disciplines of engineering and science can be offered to regular and continuing education students locally and remotely, eliminating a major obstacle to establishing a boundless and complete remote education engineering curriculum.  The concept is of particular interest for countries with sparsely populated regions and with large distances between major educational centers, which is the case for many Nordic countries.  Also, we will offer the remote lab to local colleges that can benefit from this technology to augment their engineering and science curriculum.   The technology described can also be applied to general remote instrument control in many areas of research and engineering.

 

 


Background and Pedagogical Perspectives

With the mass proliferation of the Internet, interesting possibilities have emerged for extending its use into new areas, including distance-education – a rapidly growing part of the university curricula that also encompasses the increasing continuing education market. By utilizing the Internet and the World Wide Web, the potential exists for offering courses to remote students, who can participate without other technical requirements than a personal computer and a telephone line.

 

Laboratory courses are a vital part of engineering education, but so far, lab courses have been considered impractical for distance-education. On the other hand, user-friendly, computer-controlled instrumentation is revolutionizing the way measurements are being made, and is now permitting net-based techniques to be utilized for setting up remote laboratory access. Such a remote laboratory can, for example, be used in conjunction with courses in electrical engineering, allowing remote students to gain hands-on experience in a wide range of areas from semiconductor device characterization to advanced circuit analysis. As an added benefit, this technology may offer students the opportunity to work with sophisticated equipment, of the kind they are more likely to find in an industrial setting, and which may be too expensive for most schools to purchase and maintain.

 

The basic concept and feasibility of remote system control via the Internet were investigated in two Siv.ing. (MS) student theses at NTNU [1,2]. This investigation was financed by a small grant from the DIGITALIS program at NTNU in 1997, funded by the Research Council of Norway. This investigation described a first version of the proposed system, and demonstrated its feasibility in an actual laboratory setting. Further development was pursued in 1998 in collaboration with Professor Shur at Rensselaer Polytechnic Institute (RPI) in Troy, NY, where Prof. Fjeldly was on a sabbatical leave. This work so far has been described in several publications [3-7].

 

The present version of the remote lab is dedicated to semiconductor device characterization. It includes several experiments that are performed on a microelectronic test chip, and is used as a lab module in a course on semiconductor devices and circuits at the senior or first year graduate level.  In Norway, this is a course that is presently being taught remotely from UniK to students at NTNU in Trondheim. The remote lab is planned to become a permanent part of this course. In this setting, we have already shown that on-line lab demonstrations in combination with other applications, such as circuit simulations, are fully feasible by utilizing, for example, Microsoft's NetMeeting.  By analyzing the experimental characteristics obtained from individual devices, the students can extract SPICE parameters, which can be used for simulating the behavior of circuits, for example, using our circuit simulator AIM-Spice [8]. These simulations can then be compared with experiments on simple circuits in the test chip. For the theoretical part of such a course, we use our new textbook "Introduction to Device Modeling and Circuit Simulation" [9].

 

Central issues within the proposed project are to increase the flexibility in configuring the experiments from the client side, to improve the user interface, and to increase the number of experiments.  Eventually, based on this concept, lab courses and course modules within many disciplines of engineering and science can be offered to regular and continuing education students locally and remotely, eliminating a major obstacle to establishing a boundless and complete remote education engineering curriculum.  Specifically, we aim to offer the remote lab to local colleges that can benefit from this technology to augment their engineering and science curriculum. 

 

As part of this project and in collaboration with our partners, we have written a contract with John Wiley & Sons, New York, to publish a book that describes both the technology used and the details of the various experiments.  Such a book will then serve as a lab manual for the students.  We plan to invite several Nordic and other universities to use our technology to set up separate remote lab facilities. Each participating institution will describe their own lab module in a separate chapter, according to unified guidelines.

 

 


Technical and Network Issues

The implementation of the remote lab will be based on a client/server architecture, as indicated in Fig. 1. The server can either be based on Microsoft Windows and Visual C++ or on Linux, which has several language alternatives, as indicated. An integral part of the server is the Internet communication, which allows server sockets to receive commands over the Internet. A second main component is the driver interface layer, which enables the communication between the server and the external experiment.  This communication can be done either through a GPIB card and an instrument, or directly by means of  PCI/ISA or other  cards. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Fig. 1. System configuration envisioned for the remote lab [10].

 

PCI/ISA cards can be installed directly in the expansion slots of the host PC, as an alternative to more expensive instruments. Typically, such a configuration may consist of a voltage output card and a data collection card. Controlled by the host PC via the communication buses, they have simple functionalities for outputting a single voltage and converting it into a digital signal that can be read by the host PC.  However, such configurations have limited functionalities compared to commercial instruments, and therefore require more work to develop.

 

The server will be designed to allow multi-user and multi-experiment operation. No experiment failure or errors caused by the clients should lead to malfunction of the server. If any experiment takes a very long time to finish, which suggests a failure, it will be discarded and hence will not affect the other experiments.

 

Java is indicated as one of the programming languages for the client side, since it offers the flexibility of a GUI (Graphics User Interface) design, convenient network programming, and platform independence. The client sees a pop-up window that provides interaction and communication directly with the server. Java applets provide good control, but because of Java's security structure, unsigned applets make it awkward for the client to store and present the measurement data, and to transfer them to other applications (except by "cut-and-paste").  Hence, alternatives to Java, such as JavaScript, XML (Extensible Markup Language), and WebDAV (Web-based Distributed Authoring and Versioning), that provide better utilization of the browser functionalities will be explored.  As indicated in Fig. 1, even a WAP (Wireless Application Protocol) client may be a possibility.

 

In principle, the user should be able to access the remote lab Web site using a regular Web browser, where he/she opens the client window, selects an experiment, configures the experiment, and enters parameters for the experiment (such as voltage ranges, step size, etc.). The instructions from the client are then sent to the server, which runs the experiment and returns the experimental data to the client. Depending on the technology used by the client, the experimental data can be presented to the user in different formats, as indicated in Fig. 1. 

 

 


Deliverables

The following deliverables are associated with this project:

 

1. A fully functional, state-of-the-art educational tool for performing laboratory experiments over the Internet, including:

                Software for Internet communication

                Driver software for the experimental instrumentation

                User friendly and instructive client interface

                Multi-experiment and multi-user capability

                Ability to utilize the latest in internet browser functionalities

 

2. A complete text describing the details of the remote lab, including:

                A software manual and reference

                Instructions on setting up new experiments

                Examples of experimental setups with lab manuals for students

 

3. Additional educational benefits:

                Two PhD partially completed and several MS theses to be completed within this project

                Access to lab modules will be offered to other institutions over the Internet

 

 

 


Goals and Tasks

The primary objective of this project is to complete the development of and demonstrate a fully functional remote lab system, using state-of-the-art Internet communication technology.  To achieve this goal, we propose to focus the attention on the following partial goals and tasks necessary to reach our objective. 

 

 

1.     Server. The server unit (see Fig. 1) has the dual function of communicating with the client over the Internet and controlling the experimental setup. To optimize the server function, we will explore various options regarding operating system (MS Windows, Linux) and language (ASP, Visual C++, Perl, PHP, C/C++ (cgi)).  We will also consider alternative solutions for the acquisition of experimental data, including the use of low-cost PCI/ISA cards as an alternative to high-level instrumentation. 

 

2.     User interface. It is a requirement that the client interface be user-friendly. This can be achieved, for example, by developing more instrument-like images to be displayed on the client side, in which the experimental configuration and parameters can be controlled. However, this ambition must be weighed against the speed of the system, since the use and update of graphical images from the server may cause a large transmission overhead.  On the other hand, to provide better utilization of the browser functionalities, JavaScript and XML (Extensible Markup Language), WebDAV (Web-based Distributed Authoring and Versioning), and even WAP technology will be explored as alternatives to Java.

 

3.     User control of lab configuration. Develop a switching matrix system that can be controlled by the user via the server unit. The purpose of this unit is to make physical connection to different terminals of the test objects, whereby the students easily can configure their experiments, preferably in a graphical mode. The purpose is to enhance the lab experience, to increase flexibility in selecting experiments, and to permit the introduction of additional experiments in the system.

 

4.     Speed and security. The speed has to be weighed against other concerns such as user-friendliness and the transmission of graphics. Security is always an issue in open Internet systems, and we will be considering different alternatives. Of special concern with the remote lab is to prevent hacking and the application of damaging inputs to the test circuit.

 

5.     Authentication and scheduling systems. These are practical considerations needed for an orderly execution of lab courses.  For the scheduling of sessions, we will consider the use of the MS Message Queuing application in combination with techniques based on synchronization objects, critical sections and semaphores.

 

6.     Establish remote lab sites. Remote lab sites will be established primarily at UniK/NTNU and Chalmers, but we will also actively solicit additional participants from other universities to set up such sites based on our technology and to contribute in the development of the remote lab.  Thereby, a network of participating laboratories can be established.  Hence, we foresee that a broad range of sophisticated experiments can be made accessible to students over the Internet, at a relatively modest investment for each institution. 

 

7.     Testing and evaluation.  The remote lab system and the various experiments will be tested and evaluated in relevant courses at our universities.  This will give valuable feedback on how to improve the system to enhance its pedagogical benefits.

 

8.     Documentation.  As a necessary complement to this new educational tool, we propose to publish a book that describes both the technology used and the details of the various experiments, and that also will serve as a lab manual for the students.  Each participating institution will describe their own lab module in a separate chapter, according to unified guidelines.  An interest for publishing such a book has been expressed by an international publishing house.  We also plan to publish our results at conferences and in journals, and distribute information about the project as widely as possible.

 

9.     Offer remote lab to local colleges.  Towards the end of the project, we aim to offer the remote lab to local colleges that can benefit from this technology to augment their engineering and science curriculum. 

 

10.  Other applications. We also envision applications of the present technology for general instrument control over large distances, for example, as part of an enhanced tele-medicine concept.  We also foresee it as a useful tool for scientists and engineer who may be able to investigate their samples (for example, electronic devices and circuits) remotely by sending their samples to a unique laboratory facility at some distant location.  In the laboratory, the samples may be mounted in a test facility by local staff, whereupon the investigators may perform their analyses from their home base, eliminating the need for costly and time-consuming travel.

 

 

 

 


References

 

[1] V. Kristiansen, Remotely operated experiments on electric circuits over the Internet - An implementation using Java, M. Sc. thesis, Norwegian University of Science and Technology (1997).

[2] B. Dalager, Remotely operated experiments on electric circuits over the Internet - Realizing a client/server solution, M. Sc. thesis, Norwegian University of Science and Technology (1998).

[3] AIM-Lab URL: http://nina.ecse.rpi.edu/shur/remote

[4] H. Shen, Z. Xu, B. Dalager, V. Kristiansen, Ř. Strřm, M.S. Shur, T.A. Fjeldly, J. Lü, T. Ytterdal , "Conducting Laboratory Experiments over the Internet", IEEE Trans. on Education, 42, No. 3, pp. 180-185 (1999).

[5] T.A. Fjeldly, M. S. Shur, H. Shen and T. Ytterdal, "Automated Internet Measurement Laboratory (AIM-Lab) for Engineering Education", Proceedings of 1999 Frontiers in Education Conference (FIE’99), San Juan, Puerto Rico, IEEE Catalog No. 99CH37011(C), 12a2 (1999).

[6] M. Shur, T. A. Fjeldly and H. Shen, "AIM-Lab – A System for Conducting Semiconductor Device Characterization via the Internet", late news paper at 1999 International Conference on Microelectronic Test Structures (ICMTS 1999), Göteborg, Sweden (1999).

[7] T.A. Fjeldly, M.S. Shur, H. Shen, and T. Ytterdal, "AIM-Lab: A System for ­Remote Characterization of Electronic Devices and Circuits over the Internet", Proc. 3rd  IEEE Int. Caracas Conf. on Devices, Circuits and Systems (ICCDCS-2000), Cancun, Mexico, IEEE Catalog No. 00TH8474C, pp. I43.1 –I43.6 (2000)

[8] AIM-Spice URL: http://www.aimspice.com

[9] T.A. Fjeldly, T. Ytterdal, M.  Shur, Introduction to Device Modeling and Circuit Simulation, Wiley & Sons, New York, NY (1998).

[10] K. Smith, M. Sc. thesis, UniK/University of Oslo (2000).

 

 


Short Resumés

 

Professor Tor A. Fjeldly (Electrical Engineering) received the M.S. degree from the Norwegian Institute of Technology, 1967, and the Ph.D. degree from Brown University in 1972.  He was with the Max-Planck-Institute for Solid State Physics in Stuttgart, Germany (1972-1974) and SINTEF in Trondheim, Norway (1974-1983). Since 1983, he has been with the Norwegian University of Science and Technology (NTNU).  He is presently with NTNU's Center for Technology (UniK) at Kjeller, Norway. He has been a Visiting Professor at University of Virginia (1990-1997) and at Rensselaer Polytechnic Institute (1997-). His current research interests are semiconductor device technology and circuit analysis.  He has written about 150 scientific papers, several book chapters, and is co-author of two books, including "Introduction to Device Modeling and Circuit Simulation" (Wiley & Sons, 1998). Professor Fjeldly is a Fellow of IEEE, a member of the Norwegian Academy of Technical Sciences, and a member of several other professional societies.  He is Co-Editor-in-Chief of International Journal of High-Speed Electronic and Systems, published by World Scientific Publishing Co. in Singapore.

 

Professor Kjell Jeppson (Electrical Engineering): After receiving his Ph. D. degree in solid-state electronics from Chalmers University of Technology in 1977 he became senior lecturer at the Department of Solid-State Electronics in 1978. Since 1996 he is professor at the department of Microelectronics. He spent the academic year 1973-74 with Rockwell International, Anaheim, CA and the fall semester 1985 at the Southampton University Microelectronics Centre, England. His main research interest is focused on MOS and bipolar device modeling and parameter extraction, and CMOS VLSI design. He has published several papers on MNOS nonvolatile memories, transistor modeling and parameter extraction, CMOS gate delay and hierarchical DRC of VLSI circuits. He has also authored a textbook (in Swedish) on semiconductor devices. Since 1993 he is also vice Dean, responsible for the undergraduate school of Electrical Engineering for which he has initiated the E-96 project for implementation of a new undergraduate curriculum. In 1999 he was general chairman for the International Conference on Microelectronic Test Structures (ICMTS´99).