CC BY
9
REFERENCES
1. S. Knerr, L. Personnaz, and G. Dreyfus Handwritten Digit Recognition by Neural Network with Single-Layer Training, IEEE Trans. On Neural Networks, 3, pp. 962-968, Nov. 1992.
2. A. K. Jain, J. Mao, J. and K. M. Mohiddin, Artificial Neural Network: A tutorial, IEEE Computers, pp. 31-44, March 1996.
3. W. A. Schmidt, and J. Davis, Pattern Recognition Properties of Various Feature Spaces for higher order Neural Networks, IEEE Trans. On PAMI, vol. 15, no. 8, pp. 795-801, Aug. 1993.
4. R. Lippman, An Introduction to Computing with Neural Network, IEEE ASSP Mag. Pp. 4-22, Apr. 1987.
5. A. Rajavlu, M. T. Musavi, and M. V. Shirvaikar, A Neural Network Approach to Character Recognition, Neural Networks, vol. 2, pp. 387-393, 1989.
6. D. Gabor, Communication Theory and Cybernetics, IRE Trans., CTI-4, pp. 19-31, 1954.
7. D. Gabor, W. P. Wilby, and R. Woodcock, Universal non-linear Filter Predictor Simulator which optimizes itself by a learning processes, Proc. IEE, vol. 108, Part B. no. 40, pp. 422-433, Jul. 1961.
8. J. Ahmad, "Novel Neural Network strategies based on the Gabor-Kolmogrov learning Algorithm for Pattern recognition and prediction", University of London, 1995.
9. D. E. Rumelhart, G. E. Hinton, and R. Williams, Learning internal representations by error propagation, in Parallel Distributed Processing (D. E. Rumelhart and J. L. McCelland, Eds.), Cambridge, MA:MIT Press, 1986, vol. 1, chapter 8.
10. B. Widrow and M. E. Hoff, "Adaptive Switching Circuits," IRE WESTCON CONV. Record, Part 4, pp. 96-104, 1960.
11. B. Widrow B. and M. A. Lehr, "30 Years of Adaptive Neural Networks Perceptron, MADALINE, and Backpropagation," IEEE, vol. 78, no. 9, pp. 1415-1442, Sep. 1990.
YAK 004.93
THE INTELLIGENT CIRCUIT THAT IS OPERATED BY TRANSMISSION OF IMPULSE AS SYMBOL OF ACTIVITY
Karasawa S.
Якщо вплив на датчиках представлений тдмножиною 1мпульс1в, а di'i виконуючих пристрою представлен iншою тдмножиною iмпульсiв, тодi рабоча пам'ять формуеться через дiю одночасно шляхом поеднання точок, де iснують iмпульси. 1нтелектуальна поведiнка, що описанна у логiчнiй програмi, може бути достигнута за допомогою ланцюга, що передав iмпульс як дiю.
Если воздействия на датчиках представлены подмножеством импульсов, а действия исполнительных устройств представлены другим подмножеством импульсов, тогда рабочая память формируется через действие одновременно путем соединения точек, где существуют импульсы. Интеллектуальное поведение, описанное в логическей программе, может быть достигнуто посредством цепи, которая передает импульс как действие.
If activities on sensors are represented by a subset of impulses and actions of actuators are represented by another subset of impulses, the working memory is formed through the activity concurrently by connecting the points where impulses exist. The intelligent behavior described in a flow chart can be achieved by means of the circuit that transfers an impulse as an activity.
1 INTRODUCTION
Neural engineering is an emerging discipline [1]. The greater part of artificial intelligence (AI) is researched by means of computational models. Those algorithms of the models are adapted to the digital computer. We appreciate the fruits of the software. The computer is omnipotent if the algorithm is perfect. Although the computing paradigms such as artificial neural networks [2] and connectionist models [3] are taking nature of a brain into account, the programs are difficult to form a working memory through an experience automatically.
The software met difficulties in the field of recognition if the object is things and affairs in the real world. An activity
is nonlinear and an activity is not continuous. The impulsive activity in the brain is not computational. The author as a researcher in the field of semiconductor electronics proposes the circuit where an impulse is transferred as an activity is transferred. The artificial intelligence in which the existence of impulse means occurrence of an action can be termed as hardware artificial intelligence.
Since a digital data is consist of signal on motionless state, the change of a state is represented by means of two states, i.e. one is obtained before the change, and the other is obtained after the change. There are two kinds of changes on a digital state, one is the positive impulse that causes an excitatory state, and the other is the negative impulse that inhibits an active state. A positive impulse that is the minimum unit of excitatory action is able to connect the lines at the transference, and the transmission of the positive impulse creates the working memory that makes possible to replay the same activity [4].
The action of connections makes possible to materialize the working memory. Although the definition of the working memory is difficult within the traditional concept of software, we can define the working memory as the function of a neuron, because the connections of a neuron are formed by means of activities, and the neuron decodes impulsive activities and it outputs an activity. The intelligence is a kind of activity and it is not restricted to the processing of information.
If we employ the segmentation of working memory in order to recognize things and affairs in the real world, the processing of recognition will become easy. If we divide a pattern into many pieces in order to compute, the processing of recognition becomes a jigsaw puzzle.
On the other hand, the intermittent operations of impulses make possible to carry out time-sharing of the operations. A circulating impulse in a loop is able to keep the activity, and the paralleling activities by means of transmit-
ting of impulses along plural flowcharts make possible to think while walking [5].
The activities by means of circuits driven by impulses are also possible to operate the information processing. The activity driven circuit that makes activity according to the transference of impulse is described by a flowchart on activities. The function of the flowchart is able to realize by means of impulse driven circuits. This new design concept is inspected through trial manufacturing and the results are reported in this paper.
2 PROBLEMS TO BE SOLVED BY MEANS OF
THE CIRCUITS DRIVEN BY IMPULSES AS
ACTIVITIES
We can memorize the first experience without lesson. But the computational learning model such as back propagation method [7] in neural network cannot form the circuits similar to the working memory that human possesses. It is difficult to process the signal by means of the central processing unit (CPU) that is under the construction.
The voice has an independent meaning only for the audience. The program and data are divided in a digital computer. Although the data are transferred, the machine cycle in a digital computer is not signified by the data. The meaning of signal is not given to wiring in the traditional technology. The digital signal that is independent from the media is difficult to form a working memory, for a working memory is a kind of circuit that is able to operate a function.
Another problem caused by the traditional concept of signal processing is difficulty in the processing for the recognition of things and affairs in the real world. The difficulty is caused by the structure of the signal. A digital computer deals with data adjusted to the signal processing, and there is the segmentation in a flow of signal processing. On the other hand, the things and affairs in the real world are possessed of the segmentation of itself. Since there are a great number of the combinations between two kinds of segmentations, the recognition of things and affairs in the real world becomes difficult. If we use the segmentation of the working memory for the recognition of things and affairs, the processing of recognition can be done at once.
The real activity is able to cover every information processing, but the information processing cannot cover every real activity. The serial processing of information is a part of the brain mechanisms. The traditional artificial intelligence cannot cover the function of consciousness in a brain well. The paralleling activities by means of transmitting of impulses along plural flowcharts make possible to emulate the communicating operations of the brain. The focused attention that is important for intelligent behavior will be emulated by means of the circuits designed by the concept of activity.
We can image the many new figures of a brain mechanism, if we remove the glasses of software. Although there are differences between human and machine, the interface that is designed by the concept of activity will contribute to decrease the gap between the activities in a brain and the traditional information processing.
3 THE METHOD TO ACQUIRE INTELLIGENT
CIRCUITS THROUGH ACTIVITIES
3.1 The function of a neuron
There are a huge number of neurons in a brain. A number of impulses run through numbers of circuits. The activities can be memorized in the circuit if every signal is composed of the impulse. The subset of impulsive signals is packed by means of connections on a neuron at an instant. The neuron receives many impulses and it generates an impulse by means of a biochemical reaction.
Since resting potential in a neuron is -80mV sustained by many biochemical pumps and the narrow peak of positive impulse is +40mV [6], the connections that the input terminals correspond to the points of existing positive impulses may form a decoder on the subset of impulses. A pattern of impulses is detected by the neuron formed through the same pattern of activities. This method of connections is able to form a working memory.
The subset of paralleling impulses or a series of intermittent impulses are transmitted by the working memory. The circuits for working memories are able to form automatically by making use of such activity driven system. The network of working memory is available for the cognitive device such as visual perception, or voice recognin.
3.2 Characteristics of the hardware artificial
intelligence
The working memory is operated by means of minimum unit of activity where the process of making signal is omitted. The hardware artificial intelligence is based on the activity that depends on the body. The meaning of one impulse that comes from a sensor is able to understand as a reaction of the sensor under the situation. The activity of an actuator is able to understand from the effect that exerts on the surroundings. As for the activity of nerve cell, the meaning is contained in the wiring connections.
Since the nerve cell ignites in accordance with propagation of the impulse, the operation of a nerve cell is similar to the pin of the pin-drum of a music box. The activity of one pin produces one tone of sound. The whole activities of pins become music. The sophisticated reaction in the impulse driven system is achieved by means of linked activities. These activities are represented by a subset of impulses where many wiring are needed. In such case, although an activity in a cell has the meaning that depends on the cell, we cannot understand the meaning of an impulse well when we analyze the impulse that comes from only one nerve cell.
The concept that "an impulse is the representative of an activity in the circuit for intelligent activity" is different from the traditional concept of information processing system. The activity in the real world cannot separate from the substance and it cannot not free from the real time. The information that is translated from things and affairs in the real world makes possible to free from the real time.
The information of an impulse is only the timing that belongs to the real time. The device driven by activity is designed in order to make actions in the real time. In this design concept of impulse mode operation, the existence of
impulse in a circuit means the activity in the device. An impulse is the symbol of an activity.
Although the operation of working memory in an impulse system is similar to the digital decoder, the number of the input terminal of the working memory is adjusted with the object. Although there are intermittent transmissions in the impulse driven mode system, there is not the segmentation caused by the process of producing the signal of information.
The symbol itself is not computational, but it can be handled by means of circuits. These types of intelligent circuits will be useful for the electronic device that emulates the brain mechanism. Those circuits will be useful in the field of robotics or automaton.
3.3 Differences between digital circuits and
activity driven circuits
We can replay an activity by a movie, but it is a copy on the visual image. The software is not able to function without hardware. But hardware is able to function without software. The impulse is the minimum signal on the activity, whereas the pulse is the minimum information.
The signal processing without comprehension such as a telecommunication stays within the world where activities improve expressions. The end user of a traditional signal processing is the human being. Fig.1 shows the architecture of a humanoid designed by the traditional concept of information processing.
The brain is not a mere signal processor, and it is an end user of the signal. The end user makes actions. The architecture of new machine is simple, but it operates every activity including information processing. Fig. 2 shows the animal-robot in which the digital signal processing is omitted where only activities are transmitted along plural flowcharts.
Figure 1 - Block diagram of signal processing on a humanoid designed by traditional concept (Here, the data are transferred through media)
Figure 2 - The activity driven circuits for the model of brain mechanism where only impulses are transmitted as the symbol of activities (Brain of animal that is driven by transference of activities)
A combination of positive impulse and negative impulse is given by the derivative of a pulse with respect to time. But, the meaning of "impulse in the hardware artificial intelli-
gence" is different from that of "pulse in a traditional signal processing". In a traditional signal processing, the serial arrangement of pulses or the waveform of pulse carries information. But in the intelligence circuit driven by impulses as symbol of activities, the arrangement of series of impulses does not imply the meaning, and the waveform of an impulse does not imply the meaning. The frequent generation of impulses only indicates that the activity is breaking out frequently.
The electronic circuit driven by the activity can be manufactured with the method similar to digital circuits. That is, the activity is expressed with the presence of a certain amount of electric charge. The electric charge is transferred via the route selected by the logic circuit. The circuit in which the existing of electric charge means an operation of activity is realized by means of the semiconductor elements such as CCD (Charge Coupled Device) where the lines are connected by using floatation gate MOS FET (Metal Oxide Semiconductor Field Effect Transistor).
The implementation of working memory in the impulse driven system will be extremely easy compared with traditional digital technology of PLD (Programmable Logic Device) [8]. For example, if a pattern of activities is expressed with the pattern of the lighting spots, the pattern of lighting spots will be transferred via the route selected by the membrane in which a transparent point is formed by using the pattern of photochemical reactions on a membrane. The pattern of activities can be reproduced by using this path of lights.
4 THE HADWARE ARTIFICIAL INTELLIGENCE
THAT EMULATES BRAIN MECHANISM
4.1 The loop of delay elements for a short-term
memory of activity
The function of a short-term memory for an activity is achieved by means of looped delay elements where circulation of an impulse represents a continuous activity. The excitatory state of this loop generates impulses continuously by the circulation. The plural paralleling impulses are able to excite one activity selected by the working memory that decodes the constituents.
There are activities of plural loops in the layer of constituents and that of one loop in the layered of output. The organization of activities yields activities of plural layers for the same agent. The meaning of each activity for the same agent is different. The organization of activities by means of working memories together with loops makes possible to form the frame of knowledge i.e. linkage of the activities.
The array of loops is able to operate the function of a register. The subset of impulses that should pay attention is transcribed on a register. The pattern of points activated on the register can be referred to the images those are inputted one after another successively. By using an activated image as a mark of alignment, a mapping of visual images can be carried out. In general, the consciousness that pays an attention to a certain image is realized through activities of the subset.
4.2 The organization of activities
The impulsive actions will be combined for a sophisticated activity. The subset of impulses is decoded by the working memory in which the intermittent activities are constituents. The linked working memories form a layered structure. A group of activities are decoded by means of multi-layered working memories.
The linguistic expression is transmitted in the form of subsets of impulses in a network of neurons. The faculty of language is acquired by means of the layered structure of working memories. The network of working memories is able to comprehend the meaning of speech voice and written sentence.
Since the organized structure makes use of existing circuits, the structure of extension economizes the circuit. There is compatibility in the element of component compared with the whole. The layered structure is able to represent a great many kinds things and affairs, because there are a great many kinds of combinations among many components. The things and affairs in the real world will be represented in the layered structure of working memories.
Since the output of working memory for a subset of serial activities is generated at the end of serial impulse, there are time differences among these impulses. The loop for a short-term memory is used in order to adjust the timing among plural activities. The working memory together with loops organizes sophisticated activities.
The linked activations of loops by means of a working memory make possible to provide impulses two directions. One direction is from the inputs side to the output side and the other direction is from the output side to the inputs side. The concurrent excitations of loops makes possible to form an afferent pathway and an efferent pathway. Here, the afferent pathway transfers a subset of components into the output, and the efferent pathway transfers a representative into the subset of components.
Positron Emission Tomography (PET) has enabled us to observe activities of a cerebrum [9]. The observations by PET show that the activities of cerebrum are held for a while. The working memory with loops that is driven by impulses will be able to emulate the brain activities.
4.3 The amplification of activities
The signal will be changed when the number of pulses changes in a traditional data processing where the arrangement of pulses is used as the signal. However, in the system driven by activities, the increase on the number of impulses is equal to the frequent activities. That is, to decrease the number of impulses is equal to the decrease of the frequency on activities.
Since the repetition of activity is carried out by an increase of impulses, the activity is emphasized by using of [OR] circuit for summing of impulse that was delayed. Since the repetition of an activity is diminished by a decrease of impulses, the activities are diminished by reduction of the number of impulses through AND circuit. Then, [OR] logic
circuit is able to use as an amplifier of activities. [AND] logic circuit is able to use as an attenuator of activities.
4.4 OR logic circuit for activities
Circuit of [OR] logic connection is needed where many circuits control an actuator. OR circuit outputs an impulse when an impulse inputs into one of the input terminals at least. However, OR circuit will ignite by a noise in case of many inputs. The automatic implementation of OR circuit makes such a problem.
Since the connections of AND circuit are equal to the points of existing impulses, the implementation of AND circuit can be carried out automatically. De Morgan's law can be stated as " AND of inverse of Boolean expressions is OR of the inverse of Boolean expressions". That is, OR logic on the complement of Boolean expression can be stated as "when even one pulse does not come the pulse is not output". The inhibitory (negative) impulse that suppresses excitatory (positive) impulses is used in the circuit driven by impulses.
Then, AND circuit for inverted inputs outputs an impulse when there is not an impulse. The output is used to suppress the positive impulse that is generated continuously. So. OR logic circuit is realized by using NOT, AND, and loop for generation of impulse. This circuit for OR logic connections will be able to emulate the activities on a cerebellum [10].
5 THE RESULTS; THE MANUFACTURING OF
THE ACTIVITY DRIVEN CIRCUITS
5.1 The flowchart on a hardware artificial intelligence
Signal processing in a digital system is expressed by means of a flowchart. Each step of operation can be described by a proposition of "IF A=B, then Y=X". This logic is the fundamental statement for the software. The proposition can be carried out by a working memory. The flowchart on activity driven system expresses a flow of activity. Each step of activity can be described by a proposition of "IF present state of activities is A and the inputting activities is B, then the next state of activities is Y." The active state of this device is transferred to the other state assigned by the activities of conditions sent from the monitors of itself and outer world.
The format of flowchart on activity driven system is similar to that of digital system. But, the design of an activity driven system is based on the specification of activities.
Although a state in the activity driven system is represented by the state of circuit in which an impulse exists, the procedure to design the activity driven system is similar to the design of sequential system where one step of operation can be described by the change of state. The sequential system is known as "finite state machine". The tools and parts used in such digital system will be available in the activity driven system. The operations of an activity driven system is described by means of a state diagram and a timing diagram.
5.2 Manufacturing of the activity driven circuit that is made of digital IC
5.2.1 Digital circuits for the activity driven circuit
Figure 3 - The flowchart for transference of activities
a) Delay element (that is able to use as a timer for a controller) is realized by a MM; 74LS123 (14538B). The time period is given by the T=0.7CR at 14538B;
b) Decoder is composed of an [AND] circuit; 74LS08(4081);
c) A loop for a short-term memory is made of a pair of MM;
d) An impulse in a loop is deleted by means of a 3-state gate; 4LS244(74HC244);
e) Common outputs are connected in a [OR] circuit; 74LS32(4071).
(Here, the registered number of IC is expressed as; TTL IC (c-MOS IC).
5.2.3 Trouble shooting of activity driven circuits
We manufactured several kinds of impulse driven control circuits [11], [12], [13], [14] Those circuits are the circuit for automatic vending machines, the circuit for automatic washing machine, the circuit for traffic controller, and the circuit for a sheet of short-term memories to replay the activity. Many Light Emitting Diodes (LED's) are used as monitors of the operations. These circuits functioned well.
We met some troubles. That is, the network of impulse driven circuits is delicate to the noise. Then, the chattering of switch changes the state. Moreover the trouble in activity driven circuit is transmitted by the impulse. The trouble shooting of activity driven circuit system is hard compared with the digital circuit.
Figure 4 - The element of activity driven circuit that is composed of digital IC
Fig.3 illustrates the flowchart of transference of an activity. Fig.4 shows a circuit for the impulse driven mode. The plural outputs of monitors are connected by AND circuit in order to obtain the trigger signal that changes the present state. A loop made of a pair of Monostable Multivibrators (MM's) is used for a short-term memory. Two pieces of registers (1k ohm) are used as coupling elements of a loop, where the input into the input side of register is not forced to the output of another MM.
OR circuit connected from plural outputs to one next input is used in order to excite the next state. In order to delete an activity in a loop after the transition, the output of 3-state gate is connected to input port of the loop directly. When the control signal of deletion puts in the gate terminal of 3-state gate, the output is connected to the ground, and the rounding impulse is absorbed to the ground. When the control signal of deletion does not put in the gate terminal of 3-state gate, the output is unconnected to the ground.
5.2.2 Digital IC that is used for activity driven
circuit
The activity driven circuit for a flowchart is comprised of delayed transferring element together with logic circuits as shown in Fig.4. The circuit is assembled by digital ICs listed as follows.
5.3 The semiconductor device that is used for
the activity driven circuit
The practical activity driven device must be manufactured as a semiconductor device. A quantity of electric charge is able to represent an activity. Since the function of a flowchart is able to realize by means of a charge transfer device, the MOS technology will be useful in the field of impulse device. A dynamic Random Access Memory (RAM) consists of a capacitance where the charge is stored. The charge in the Charge-Coupled Device (CCD) can be moved along the surface of a semiconductor under the application of a proper sequence of voltage pulses, the charge packet can be transferred in a controlled manner. CCD is distinct from conventional device where current and voltage levels are generally used. However, CCD is able to realize digital logic circuits such as OR, AND, INVRERT [15]. A charge-sensing amplifier nondestructively generates the same amount of another electric charge in one location and this information is able to transmit another location [16]. A memory element of a floating-gate avalanche-injection MOS memory (FAMOS) is fabricated as a dual-stacked poly-silicon structure in which the bottom gate is floating. The floating gate is isolated by thin oxide, and it provides the area for memorizing electrons. In EEPROM (Electrically Erasable Programmable Read-Only Memory), a circuit called a charge pump is used to generate the necessary programming voltage from the standard 5V, and the internal connection by means of the floating gate is provided to reverse the electron injection phenomenon [17]. Then, a pattern of activities can be copied to the pattern of
connections where the pre-injected electrons are removed. Since a positive impulse represents an activity, the function of a negative impulse is given by means of the circuit that absorbs the positive impulse.
6 SUMMARY
The characteristics on the activity driven circuits and the design tools are reported in this paper. We made several activity driven circuits by using digital IC's. The manufactured circuits functioned well. But, now, those circuits are not much superior than the digital circuits. In principle, the activity driven circuit can be designed to form through the experience. The function of flowchart that is designed by the concept of activity will be transferred into a charge transfer circuits. It is possible to fabricate activity driven circuits as a semiconductor integrated circuit. Such activity driven circuit will be superior to today's digital system.
The trial manufacturing of activity driven circuit without automatic implementation is the step to realize the activity driven system. The manufactured circuit operates the designed function. These experiences teach us the importance of timing on the operation.
As another fruit of this research, the concept of activity helps to manufacture the circuit model for the brain mechanism. The impulsive actions are combined for a sophisticated activity. The group of activities is decoded by means of a working memory. A loop is able to keep an activity. The consciousness is intermittently interlocked to the other activity. These linkages of the short-term memories make possible to explain the activities of a brain such as linguistic activity. The intermittent transmission of activity will provide the deeper understanding of the brain mechanism that has relations to the artificial intelligence.
Although the concept of activity driven circuit is based on the activity in the real world, the circuit is a model. The author hopes that this new concept of hardware artificial intelligence will contribute to developments of the technology of intelligence.
ACKNOWLEDGMENT
The author is grateful to them who manufactured the impulse driven circuits as graduation studies. Mr. Y. Uchi-umi made a control circuit for automatic vending machine [13], Mr. Y. Fujiwara made a control circuit for automatic
washing machine [14], and Mr. Y. Monma made a sheet of short-term memories. [15]. The author would like to thank Prof. Takeshi Kurobane and Prof. Kimio Shibayama at Tohoku Univ. for their advices and encouragements as author's former teachers.
REFERENCES
1. Proc. IEEE, Vol.89, No7, Edited by M. Akay, Special issue on Neural Engineering: Merging Engineering and Neuroscience pp. 991.
2. Schalkoff R.J., Artificial Neural Networks, The McGraw-Hill Co. Inc. 1997.
3. Levy J.P., Bairaktaris D, Bullinaria J.A., Cairns P., Connection-ist Models of Memory and Language, UCL Press Limited, 1995.
4. Karasawa S., The strategy of impulse driven working memory for visual perception, Proc. of Inter. Conf. on Imaging Science, Systems, and Technology, June 24-27, Las Vegas, Vol.2, pp. 729-735, 2002.
5. Karasawa S., Impulse recurrent loops for short-term memory which merges with experience and long-term memory, Proc. of 3rd Int. Conf. on Cognitive and Neural systems, pp. 36, Boston Univ., May 26-29, 1999.
6. Nicholls J.G., Martin A.R., Wallace B.G., From Neuron to Brain, pp. 12, Sinauer Associates, Inc. Publishers, 1992.
7. Karasawa S., Decoder-for-neuron translation models for brain mechanism, Research report, Miyagi National College of Technology, pp. 19-25, 2000.
8. Carter J.W., Digital Designing with Programmable Logic Devices, Prentice-Hall, Inc, 1997.
9. Honjo I., Language Viewed from the Brain, Karger, ISBN 38055-6789-8, 1999.
10. Karasawa S., Oomori J., Impulse circuits for a distributed control inspired by the neuroanatomical structure of a cerebellum, Intelligent engineering systems through artificial neural networks, Vol.10, pp. 171-190, ASME press, New York, 2000.
11. Karasawa S., Monma Y., Fujiwara Y., and Uchiumi Y., Semiconductor devices for intelligent circuit that is operated by means of transferring impulses, Academics lecture meeting, Northeast branch of Japan Society of Applied Physics, pp.2627. Dec. 5, 2002.
12. Uchiumi Y., Manufacturing of an impulse driven circuit for the control of vending machine, Graduation Thesis at Department of Electrical Engineering, Miyagi National College of Technology, March 4, 2003.
13. Fujiwara Y., Manufacturing of an impulse driven circuit for the control of automatic washing machine, Graduation Thesis at Department of Electrical Engineering, Miyagi National College of Technology, March 4, 2003.
14. Monma Y., Manufacturing of a sheet of impulse driven short-term memories, Graduation Thesis at Department of Electrical Engineering, Miyagi National College of Technology, March 4, 2003.
15. Zimmerman T.A., Allen R.A., and Jacobs R.W., Digital chargecoupled logic, IEEE Journal of solid-state circuits, Vol.SC-12, No. 5, pp. 473-485, Oct. 1977.
16. Handy R.J., Use of CCD in the Development of digital logic, IEEE Trans. ED, Vol.24, No.8, pp. 1049-1061, 1977.
17. Geiger R.L., Allen P.E., and Strader N.R., VLSI design techniques for analog and digital circuits, pp. 826, McGraw Hill Publishing Co. 1990.
YAK 681.32:007.52
SOLVING OF BOUNDARY VALUE PROBLEMS FOR MATHEMATICAL PHYSICS EQUATIONS IN CELLULAR NEURAL NETWORKS
M.A.Novotarskiy
Cmamma npuceanena eueneunn çacmocyeaHHH MOKaMbHO-acunxpoHHux Memodie do eupiweHHH Kpaûoeux çadan MameMa-munHOÏ ôiçuêu Ha KëimKoeux HeûpoHHux Mepeœax. PoçiMMHymi apximeêmypu mpaduuiÛHoï aHaëoioeoï ma ducêpemHoï HeupoH-
Hux Mepeœ i apximeKmypa ôaïamomapoeoï KëimKoeoï HeûpoHHoï Mepeœi 3 ôaïamocimKoeuM çiëadxyeaHHMM. npedcmaeëeHuû aëzopumM acuHxpoHHoio ôyHKuioHyeaHHH dëa ôazamowapoeoï KëimKoeoï HeûpoHHoï Mepeœi.
