This paper describes a noncontact IC-card-type automatic passenger gate system. First, the paper briefly presents a review of the technological aspects and then describes the functions and characteristics of the card, transmitter, main unit of the gate, and card checker which constitute the system. In particular, the following issues resolved by the authors are described in detail: (1) The development of insensitive zones in the communication from card to transmitter has been eliminated through the use of subcarriers. (2) Interference during the operation of multiple transmitters has been resolved by reducing the difference among carrier frequencies, using different transmitter. (3) Intercard interference has been eliminated through the use of a polling communications system which can accommodate card holders of multiple cards. (4) The content of the card can be checked through the use of the card checker. Using a highly refined automatic passenger gate system incorporating the functions described above, experiments were performed at an actual station. Processing capacity approximately 20% higher than that available with conventional magnetic card systems was obtained, and the applicability of the new card system to practical situations has been confirmed. In the last section, issues to be resolved in the future popularization of the system are summarized.

View full abstract

In this paper we propose a new train communications network (TCN) system for processing instrumentation data collected between vehicles that have a very rapid data rate.
Now there are many TCN for communication available to control devices in a train, which have a data rate of 10k-1Mbps (bits per second). But we need a faster data rate for instumentation data, when we conduct a track test of MAGLEV (Magnetic Levitation Vehicle). In this case, we handle several hundreds of sensors on the train and collect synchronized data for structure analysis and vehicle dynamics analysis after the test.
We have developed a new TCN system for collecting synchronized instrumentation data consisting of a huge digital data-recorder and an optical-fiber train bus. This system treats a data rate of 32Mbps. And we have developed all devices, protocol, optical-module, optical-connectors and so on.

View full abstract

This paper discusses the design of PRIME (Progressive interlocking by utilizing microelectronics) which allows the signal engineer to modify the function of interlocking system, controls the train route safely within a station area, while operating time.
This system is a railway signal control system, composed by bases of 32 bits fail-safe computer. The modification of function of interlocking system while operating time is achieved by on-line loading of modified data, data transmission between mutual fail-safe computers, memory separation of each type data, fixing of data size by adoption of matrix structured data, checking of non-modifying part by supporting tool.
By adopting the above design, the modification work of function which have been done only in the night time, after finishing train operation, can be performed while train operation time. It makes rapid progress of improvement work and maintenance work of interlocking system.

View full abstract

We have developed a multimedia information system for railroad passengers, that can retrieve graphics and full-color images as well as alphanumeric data. This system consists of a host computer, user terminals, and workstations for creating and editing information pages. These components are linked through a high-speed Ethernet LAN. The user terminals requires high-speed response, so JPEG compression is used in full-color images to reduce the time required to transmit images from the host computer.
In September 1993, experimental user terminals were set up in Kintetsu Namba Station, Uehonmachi Station and Abenobashi Station. The tourist information on the Shima Spain Village and Ise-Shima National Park provided by these user terminals has been highly evaluated by users.

View full abstract

In this paper development of a scheme design and verification tool based on software design diversity for the newest solid-state interlocking system named PRIME, is introduced. Interlocking systems need different specifications on each stations. The interlocking tables form most of that specification. By utilizing this tool, the signal engineer is able to manage large number of data easily and efficiently, and compiles interlocking data from the interlocking tables for PRIME, Interlocking function which is reqired fail-safe control, is depend on interlocking data. Therefore fault avoidance of interlocking data is indispensable. It is for this reason that we utilized software design diversity for construction of this tool.
We have four phases of the design diversity in the software life cycle. Phase-1: Diversity data input of interlocking table and geographical data. Phase-2: Decompile of interlocking database. Phase-3: Interlocking data compile and interlocking functional test. Phase-4: Diversity compile for track remodeling support data.

View full abstract

On the railroad line, the trains are not running on schedule when a traffic obstruction or accident occurs. In these cases, the operator adjusts the timetable and he tries to dissolve the delay and confusions on the schedule as soon as possible. The train schedule adjustment is a very difficult task because it depends on empirical and intuitive expertise. This paper introduces a knowledge-based system approach to construct a decision support system for adjusting train schedule. Event tracking based diagram simulation method is proposed to make a high-speed diagram simulation of future arrival and departure times of trains. A cooperative man-machine interaction model is also introduced to develop an automatic schedule adjustment planning module.

View full abstract

High-dencity traffic operation is required for urban commutation lines. To solve this problem, many improvements have been made on the field of signal control. It seems necessary to organically connect Signal•Train•Operation to realize economical and safty-proven high-dencity traffic operation.

View full abstract

The processing type of a control system is compared to a combinational circuit model (CCM) or a sequential circuit model (SCM). In the CCM the output is defined uniquely by a given input set. But the output from an SCM-type system is not able to be defined uniquely by a given input set, but it depends on a system status at given time and a given input set. Then the SCM-type system can perform multiple functions depending on system status using even a little input set effectively. If there is a method which can transform the system from CCM into SCM, we can improve economically many conventional low level CCM-type systems. We propose an idea named Data-Oriented control Method (DOME) which realizes a system change from CCM to SCM. In this paper, it is illustrated how by mounting an intelligent processing equipment on the cab and relying on the DOME concept, any existing CCM type signalling system can be modified to a new integrated train protection system (ITP) which can perform sophisticated functions of train protection economically. And an ITP's preceding system developed on conventional low-level ATS ground equipment is introduced.

View full abstract

As one of the basic research areas for the improvement of magnetic shielding effects and the fabrication of largesized cylinders, the present paper shows how the shielding is improved by superimposing a single- or doublemagnetic cylinder such as a soft iron cylinder over a high-critical-temperature (high-T_c) superconducting cylinder such as a Y-Ba-Cu-O (YBCO) cylinder. The value of the maximum shielded magnetic field (B_s) for the superimposed cylinder by using a double-magnetic cylinder was found about 50 times greater than that of a superconducting YBCO cylinder. The values of B_s are systematically measured by changing the lengths of single- and double-soft-iron cylinders at a temperature of 77.4K, and are analyzed by using several different models, with an important criterion being derived for designing an effective shield of the vessel. In addition, it was found that the experimental values of B_s agree well with theoretical values.

View full abstract

The requirements for the LC tuning circuit of digital communication equipments, the authors started developing in 1963, were so tight that a resonant frequency stability, Δf/f₀=±0.274% (1.4 to 9.2V at -30 to 65°C for 20years) was more tight by one order of magnitude than that of conventional analogue ones. To satisfy these requirements, we studied inductors and revealed factors of their characteristics changes.
First, a ferrite pot core was selected because of its stable characteristics achieved by those days' rapid progress using study results of disaccommodation and so on in the ferrite field. The effective permeability was hold down to 10 to improve the stability. The core size was studied using Mn-Zn and Ni-Zn that could be available in the operating frequency. A 18/14-10 Ni-Zn ferrite pot core was revealed to be most suitable one including Q-characteristics. Aging characteristics was found to satisfy the requirement ΔL/L=-0.05±0.14% from 20-year measurement. This result verifies that 20-year aging characteristics of inductors can be preestimated with high accuracy from short-term data.

View full abstract

A buried electrode image sensor with a neutral photo-electric conversion area which is connected to a common drain was proposed to relize high photo-sensitivity wide, dymanamic range and low smear. The performance of the device with 1, 000×1, 000 cell/cm² was simulated by a numerical analysis which solves simultaneous continuity equations for potential and electron and hole currents. The analysis indicated that the storable signal charge reaches 1.8×10⁵hole/cell with the dynamic range and smear of 83.4dB's and -136dB's respectively.

View full abstract

Petri nets have been widely recognized as useful models for concurrent discrete event systems such as parallel processing systems, distributed systems, communication protocols, and sequential control systems. In the restricted communication protocol, there were many modeling methods and analyses.
In this paper, we consider the communication protocols such as transport layer, session layer, presentation layer, application layer, etc. and the problem of modeling and verification for the state transition table of communication protocols by using Petri nets. We derive the modeling method by using Petri nets and investigate the structural properties of proposed models. Then, we show the usefulness of this model.

View full abstract

This paper is a first step to develop a new control system which has better human like abilities than the conventional control system, and proposes a new method of modelling of brain's function distribution. By comparison with the neuro network which is structured homogeniously, proposed model is based on the petri net which is composed of state and transition and in the petri net, it is introduced learning and self-organizing capabilities which result from modified back-propagation and Hebb like method. But, it is the fundamental different point from the neuro network that learning and self-organizing of the proposed method are carried out on the only network pass of the token transfer. From simulation results, it has been cleared that proposed model can realize brain like function distribution, and therefore, in order to obtain strong non-linear functions, proposed model is superior to the neuro network.

View full abstract

The traditional frequency stabilization method of a semiconductor laser using the Rb absorption line as a frequency reference is difficult under direct frequency shift keying (FSK). Moreover, if the signal is modulated by different duty ratio conditions of pulse signals, the frequency-locking point moves depending on the duty ratio. This moving frequency-locking point degrades the frequency stability. The peak-hold method, which has been devised for improving frequency stability under direct FSK, is clearly demonstrated here and its stability becomes better than that of the traditional method. However, if the direct modulation signal is not a good FSK signal or if the only one component of two FSK signals succeeds, the peak-hold method does not operate well enough. In this work, the coding system which keeps two frequency components in the signal and the waveform modifying circuit which makes a good FSK signal are introduced. The obtained frequecy stability is confirmed unchanged in several modulation conditions.

View full abstract

We applied Kohonen's SOMs (Self Organizing feature Map) in the color matching problem in the combination of fifteen buttons and cloths. The chromaticity of cloth's color is measured by the method based on CIE-Lab. Each distribution of fifteen buttons is represented and formed by three SOMs on each two dimensional projected plane. From the observation of these SOMs, we confirm that each distribution of fifteen buttons has many mutually overlapping regions. In this situation, we propose a method of classification that have characteristics of ranking by adaptation in each button's SOM, and test the performance of our method using practical data. From the results, we obtained over 87% correct judgement, which is defined as correspondence of teacher's class with one of the class in the best three by our method. Also our method and LVQ obtained almost the same recognition rate about the best answer.

View full abstract

Job-shop scheduling is not easy to be solved analytically, therefore, it is ordinarily solved by computer simulation using heuristic dispatching rules. The SLACK rule to give the priority to the job having the shortest margin to its due-date is effective to keep the due-dates. The margin is simply calculated by subtraction of remained operation-time from the time to the due-date. However, actual margins become shorter than the calculated ones due to the confliction between jobs.
In the preceding paper of authors^{(16, 17)}, a method to estimate rather accurate margins using two neural networks was proposed in which a three-layer neural network (numbers of neurons; the first layer-the second-the third: 10-20-1) estimated the margins and the other three-layer neural network having the composition of 11-20-1 judged the accuracy of the estimation because the reliability of the estimation was low. The estimated margins were used for scheduling using the SLACK rule instead of ones calculated by the ordinary method when its reliability was judged high. In this paper, a method using only a neural network for the estimation of which the organization is more adequately adjusted is proposed. In the proposed method, the neural network having four layers with fewer neurons as 8-8-8-1 learns the margins to due-dates from schedules made by using the ordinary SLACK rule, as human schedulers do. It is verified by scheduling simulation that the proposed method is effective to improve the maximum lateness to due-dates on job-shop schedules. Furthermore, re-learning of the neural network based on schedules made by the above method is revealed to be effective for further improvement of the schedules.

View full abstract