3. Eisenstadt, M., Domingue, J., Rajan, T. & Motta, E. (1990) Visual Knowledge Engineering. In IEEE Transactions on Software Engineering, Vol.16, No.10, pp.1164-1177.
4. Fensel, D. (2001) Ontologies: A Silver Bullet foe Knowledge Management and Electronic Commerce. Springer.
5. Gavrilova, T., Voinov, A. (1998) Work in Progress: Visual Specification of Knowledge Bases // Lecture Notes in Artificial Intelligence 1416 “Tasks and Methods in Applied Artificial Intelligence”, A.P.del Pobil, J.Mira, M.Ali (Eds), Springer, pp. 717-726.
6. Gavrilova, T.A., Voinov, A., Vasilyeva E. (1999) Visual Knowledge Engineering as a Cognitive Tool / Proc. of Int. Conf.
on Artificial and Natural Networks IWANN’99, Spain, Benicassim. - pp.123-128.
7. Gavrilova, T. (2003) Teaching via Using Ontological Engineering // Proceedings of XI Int. Conf. “Powerful ICT for Teaching and Learning” PEG-2003, St.Petersburg, p. 23-26.
8. Gavrilova, T., Kurochkin M., Veremiev V. (2004) Teaching Strategies and Ontologies for E-learning // Int. Journal “Information Theories and Applications”, vol.11, N1, pp.35-42.
9. Gmez-Prez, A., Fernndez-Lpez, M., Corcho, O. (2004) Ontological Engineering with examples from the areas of Knowledge Management, e-Commerce and the Semantic Web, Springer.
10. Gruber, T. (1993) A translation approach to portable ontology specifications. Knowledge Acquisition, Vol. 5, pp. 199-220.
11. Guarino, N. & Giaretta, P. (1998) Ontologies and Knowledge Bases: Towards a Terminological Clarification. In Towards Very Large Knowledge Bases: Knowledge Building & Knowledge Sharing, IOS Press, pp.25- 32.
12. Guarino, N., Welty, C. (2000) A Formal Ontology of Properties. In R. Dieng and O. Corby (eds.), Knowledge Engineering and Knowledge Management: Methods, Models and Tools. 12th International Conference, EKAW2000.
Springer Verlag, pp. 97-112.
13. Jasper, R. and Uschold, M (1999). A Framework for Understanding and Classifying Ontology Applications. In 12th Workshop on Knowledge Acquisition Modeling and Management KAW'99.
14. Jonassen, D.H. (1998) Designing constructivist learning environments. In Instructional design models and strategies (Reigeluth, C.M. (Ed), 2nd ed., Lawrence Eribaum, Mahwah, NJ.
15. Miller, G. (1956) The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Proccessing Information. The Psychological Review, vol. 63, pp. 81-97.
16. Mizogushi, R. and Bourdeau J. (2000), Using Ontological Engineering to Overcome Common AI-ED Problems.
International Journal of Artificial Intelligence in Education, v. 11, pp.1-12.
17. Neches, et al, (1991) Enabling Technology for Knowledge Sharing. AI Magazin, Winter, pp.36- 56.
18. OilEd, Bechhofer, S. and Ng G. Accessed from http://oiled.man.ac.uk/ at December 7, 2004.
19. OntoEdit, AIFB, University Karlsruhe Accessed from http://www.ontoknowledge.org/tools/ at December 07, 20. OntoBroker, Accessed from http://ontobroker.aifb.uni-karlsruhe.de/ at December 7, 2004.
21. Ontolingua, Stanford University. Accessed from http:// www.ksl.stanford.edu/ software/ontolingua/ at December 7, 2004.
XII-th International Conference "Knowledge - Dialogue - Solution" 22. Protg, Stanford Medical Informatics. Accessed from http://protege.stanford.edu/ at December 07, 2004.
23. Sowa, J. F. (1984) Conceptual Structures: Information Processing in Mind and Machine. Addison-Wesley, Reading, Massachusetts.
24. Scott, A., Clayton, J.E. & Gibson E.L. (1994) A Practical Guide to Knowledge Acquisition, Addison-Wesley.
25. Swartout, B., Patil, R., Knight, K. & Russ, T. (1997) Toward Distributed Use of Large-Scale Ontologies. In Ontological Engineering, AAAI- 97 Spring Symposium Series, pp.138- 148.
28. Uschold, M., Gruninger M (1996). "Ontologies: Principles Methods and Applications", Knowledge Engineering Review,vol1, N1.
29. Wielinga, B., Schreiber, G. & Breuker J. (1992) A Modelling Approach to Knowledge Engineering. In Knowledge Acquisition, 4 (1), Special Issue, pp.23-39.
30. Werthheimer, M. (1959) Productive Thinking, HarperCollins.
Authors' Information Tatiana Gavrilova – Professor Saint-Petersburg State Polytechnical University, Politechnicheskaya 29, 195251, St. Petersburg, Russia, firstname.lastname@example.org Seppo Puuronen – Professor of University of Jyvskyl, P.O. Box 35, FIN-40014, Jyvskylnyliopisto, FINLAND, Finland, email@example.com CONNECTION OF NETWORK SENSORS TO DISTRIBUTED INFORMATION MEASUREMENT AND CONTROL SYSTEM FOR EDUCATION AND RESEARCH Sergey Kiprushkin, Sergey Kurskov, Eugene Sukharev Abstract: The development of the distributed information measurement and control system for optical spectral research of particle beam and plasma objects and the execution of laboratory works on Physics and Engineering Department of Petrozavodsk State University are described. At the hardware level the system is represented by a complex of the automated workplaces joined into computer network. The key element of the system is the communication server, which supports the multi-user mode and distributes resources among clients, monitors the system and provides secure access. Other system components are formed by equipment servers (CАМАC and GPIB servers, a server for the access to microcontrollers MCS-196 and others) and the client programs that carry out data acquisition, accumulation and processing and management of the course of the experiment as well. In this work the designed by the authors network interface is discussed. The interface provides the connection of measuring and executive devices to the distributed information measurement and control system via Ethernet.
This interface allows controlling of experimental parameters by use of digital devices, monitoring of experiment parameters by polling of analog and digital sensors. The device firmware is written in assembler language and includes libraries for Ethernet-, IP-, TCP- и UDP-packets forming.
Keywords: distributed information measurement and control system, network sensors, Ethernet Interface, clientserver technology, distance education.
348 AI and Education Introduction Up-to-date systems of experiment automation are recently built on modules of software-controlled devices or digital measurement hardware, connected to interface bus. In all cases, hardware is connected to computer with interface device.
Integration of distributed system with remote sensors is more efficient, when the network interface used. It can be built on network chip and microcontroller. By using Ethernet interface, it is possible to connect different digital and analog devices, and the connection with servers will be based on TCP/IP networks.
The goal of this work is to develop a network interface for connecting remote sensors and execution units to distributed information measurement and control system for physical experiments.
The Distributed Information Measurement and Control System The distributed information measurement system (Figure 1) is based on client/server technology and works in the nets on the basis of TCP/IP protocol stack [Gavrilov et al, 2003] - [Kiprushkin et al, 2005].
The system provides the remote access to information and hardware resources of automation equipped working places. The access to physical equipment is provided by the equipment servers (CAMAC server, GPIB server, the server of access to the MCS-196 microcontrollers and others). The communication server integrates the whole distributed systems. Its functions are: communication with user, system monitoring, security, and proper distribution of resources in multiuser mode.
The experiment process is determined and conducted by client software running on a client computer. It is necessary to emphasize, that the managing experiment software are operated not on the remote computer (as when using Web technologies) [Barrie et al, 1996], [Зимин и др., 2002], but on the user one, connected to the system via global network.
The communication server, the equipment servers, and the client software are implemented as Java applications.
The data exchange between them is based on TCP stream sockets provided by java.net package, which is included into Java API standard package. The methods of using the input-output ports for the access to the interface controllers are written in C programming language.
Figure 1. The scheme of the distributed information measurement and control system Ethernet Interface There are many specialized processors (network chips) designed for communication over networks. But it is necessary to create a central command unit, which will communicate with devices and control the network chip.
This unit can be built on microcontroller.
XII-th International Conference "Knowledge - Dialogue - Solution" Choosing network chip, it is necessary to take into account the physical environment and the required transmission rate of data. For communicating over 10 Mbps network, based on twisted pair, Realtek RTL8019AS processor was selected. This chip is compatible with ISA personal computer interface by timings, data and address signals.
By emulating ISA bus with microcontroller, it is possible to gain proper network chip functioning. Atmel AVR microcontrollers are good choice to implement this idea. They perform each instruction per one clock period, so their performance is 16 MIPS for 16 MHz clock rate. This performance is enough for ISA emulation.
Atmel ATMega8535 was used in the described device. It has 8-channel 10-bit ADC, 8 Kb Flash ROM, 512 b EEPROM, 512 b RAM, pulse-width modulator and analog comparator.
The main logic of the device functioning is described below. When the device is turned on, microcontroller firmware program is initiated. By sending the RESETDRV signal, the network chip is resetting. Then microcontroller configures the network chip. Configuration can be made in accordance to the desired aim of operation: e.g. reading data from measurement device and sending these data to specified network address.
The operation modes are:
• Control of the experiment execution through digital or analog devices (relays, step motors, gas injectors, etc.) • Control of some parameters by polling analog and digital sensors (pressure, temperature, and optical sensors, atom beam sensor, etc.).
• Notify of parameter value, registering by measurement device.
Software It is possible to present information flow as follows.
Analog value is converted into binary code by ADC. This very value must be received "on the other end of wire" for placing into database. The result is put into TCP packet. TCP protocol provides the reliable transmission of the messages between remote application processes. Then the IP datagram is formed from TCP packet (the level of the internetworking) and is sent to the bottom level – a network interface level.
Figure 2. Set of the program modules and sequence of the transmission of the frame, received from remote device by PC Protocols of this level must provide the integration into global network: TCP/IP network must have a facility of the integration into any other networks, which doesn’t depend on internal technology of data communication in these networks. Hence, this level is impossible to define once and for all. An interface facility must be designed for every communication technology. IP-frames to Ethernet encapsulation protocol pertains to such interface facility.
Encapsulation of IP-package into Ethernet-frames is described in RFC1042. Then Ethernet-frame is sent via communication media.
350 AI and Education The other side receives the frame and performs the re-conversion by correspondent server software. Processing of the frames encapsulation doesn’t take much processor time in personal computer. But microcontroller has smaller speed and less memory. That’s why it is very important to solve this task by means of optimized algorithm and assembler language.
Set of the program modules and the sequence of the transmission of the frame, received from remote device by personal computer, are shown on Figure 2.
It is possible to use complete 3rd-party libraries and functions for TCP/IP implementation, because the software requires standard interactions only. These libraries are presented in all up-to-date programming environments (Java,.NET, Visual C++, Delphi, LabVIEW, etc.).
There is a question that needs to be answered by a developer of software: what level of TCP/IP stack must be implemented. This choice depends on software, used in personal computer, dataflow and processing speed of microcontroller, as well as requirements of reliability of information delivery.
The more preferred way is to use the lowest possible level of TCP/IP stack, sending data in Ethernet-, or IP-frames.
E.g. measured values of remote temperature sensor can be packed into Ethernet frames directly if qualification of the developer is sufficient to use the Ethernet level. But if the LabVIEW used, then you need to use all modules for package framing (from ethernet.asm to tcp.asm) on microcontroller side. Using the Java language for writing client applications also superimposes the restriction: when TCP-socket is used, you need to use the tcp.asm library in microcontroller. If you use UDP (unreliable delivery protocol), you must encapsulate messages with udp.asm library. This library works at transport level of TCP/IP stack that obliges to use as well as all underlayed protocols. Firmware, designed for the described device, is written on assembler language and includes Ethernet-, IP-, TCP- and UDP-package libraries.
The described device can be used in other networks, based on other protocols, but in this case it is necessary to develop the libraries for generation of correspondent frames.
Internet does not give any warranty for time of package delivery. This reason limits the use of the device in system that imposes hard time restrictions of information delivery. This feature can be eliminated by using a special network, used for undertaking the physical experiment only.
Материалы этого сайта размещены для ознакомления, все права принадлежат их авторам.
Если Вы не согласны с тем, что Ваш материал размещён на этом сайте, пожалуйста, напишите нам, мы в течении 1-2 рабочих дней удалим его.