IEEJ Transactions on Electronics, Information and Systems
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
Volume 132 , Issue 1
Showing 1-29 articles out of 29 articles from the selected issue
Special Issue on “Statistical Design and Application of Stochastic Optimization and Machine Learning”
Preface
Special Issue Review
  • Hiroshi Someya, Hisashi Handa, Seiichi Koakutsu
    2012 Volume 132 Issue 1 Pages 2-5
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This article presents a review of recent advances in stochastic optimization algorithms. Novel algorithms achieving highly adaptive and efficient searches, theoretical analyses to deepen our understanding of search behavior, successful implementation on parallel computers, attempts to build benchmark suites for industrial use, and techniques applied to real-world problems are included. A list of resources is provided.
    Download PDF (609K)
  • Seiichi Ozawa
    2012 Volume 132 Issue 1 Pages 6-13
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Along with the development of the network technology and high-performance small devices such as surveillance cameras and smart phones, various kinds of multimodal information (texts, images, sound, etc.) are captured real-time and shared among systems through networks. Such information is given to a system as a stream of data. In a person identification system based on face recognition, for example, image frames of a face are captured by a video camera and given to the system for an identification purpose. Those face images are considered as a stream of data. Therefore, in order to identify a person more accurately under realistic environments, a high-performance feature extraction method for streaming data, which can be autonomously adapted to the change of data distributions, is solicited. In this review paper, we discuss a recent trend on online feature extraction for streaming data. There have been proposed a variety of feature extraction methods for streaming data recently. Due to the space limitation, we here focus on the incremental principal component analysis.
    Download PDF (479K)
Special Issue Paper
<Systems, Instrument, Control>
  • Mai Kiuchi, Takashi Onoda
    2012 Volume 132 Issue 1 Pages 14-20
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Intrusion detection is considered effective in control systems. Sequences of the control application behavior observed in the communication, such as the order of the control device to be controlled, are important in control systems. However, most intrusion detection systems do not effectively reflect sequences in the application layer into the detection rules. In our previous work, we considered utilizing sequences for intrusion detection in control systems, and demonstrated the usefulness of sequences for intrusion detection. However, manually writing the detection rules for a large system can be difficult, so using machine learning methods becomes feasible. Also, in the case of control systems, there have been very few observed cyber attacks, so we have very little knowledge of the attack data that should be used to train the intrusion detection system. In this paper, we use an approach that combines CRF (Conditional Random Field) considering the sequence of the system, thus able to reflect the characteristics of control system sequences into the intrusion detection system, and also does not need the knowledge of attack data to construct the detection rules.
    Download PDF (597K)
  • Cheng-you Cui, Ji-sun Shin, Michio Miyazaki, Hee-hyol Lee
    2012 Volume 132 Issue 1 Pages 21-31
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Real-time traffic signal control is an integral part of urban traffic control system. It can control traffic signals online according to variation of traffic flow. In this paper, we propose a new method for the real-time traffic signal control system. The system uses a Cellular Automaton model and a Bayesian Network model to predict probabilistic distributions of standing vehicles, and uses a Particle Swarm Optimization method to calculate optimal traffic signals. A simulation based on real traffic data was carried out to show the effectiveness of the proposed real-time traffic signal control system CAPSOBN using a micro traffic simulator.
    Download PDF (2149K)
<Softcomputing, Learning>
  • Xiu Wei, Wenqiang Zhang, Wei Weng, Shigeru Fujimura
    2012 Volume 132 Issue 1 Pages 32-41
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This paper proposed a multi-objective local search procedure (MOLS). It is combined with NSGA-II for solving bi-criteria PFSP with the objectives of minimizing makespan and maximum tardiness. Utilizing the properties of active blocks for flow shop scheduling problem, neighborhood structures MOINS (multi-objective insertion) and MOEXC (multi-objective exchange) are designed in order to improve efficiency of perturbation. Any perturbation based on MOINS and MOEXC takes effect on different criteria simultaneously. The original idea of MOLS is systematic change neighborhoods in the local search procedure. The search direction of MOLS on an individual is naturally guided by interaction of MOINS and MOEXC. Moreover, there is no need to set parameters in MOLS. The MOLS combined with popular multi-objective evolutionary algorithm NSGA-II (Non-dominated Sorting Genetic Algorithm-II) is called as “NSGA-II-MOLS”. To illustrate the efficacy of proposed approach, four different scaled problems are used to test performance of NSGA-II-MOLS. The numerous comparisons show efficacy of NSGA-II-MOLS is better than most of algorithms even with the same number of individual evaluations and parameters setting.
    Download PDF (1405K)
  • Kazuaki Masuda, Kazuaki Harada, Kenzo Kurihara
    2012 Volume 132 Issue 1 Pages 42-52
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This paper proposes a general methodology of multi-objective optimization based on the combined use of scalarization and evolutionary computation approaches. Mathematically, it is guaranteed that a Pareto optimal solution of a multi-objective optimization problem (MOP) can be found by minimizing the corresponding augmented Tchebysheff scalarized function. In this way, different Pareto optimal solutions can be obtained by solving different single-objective optimization problems (SOPs) with different weight for scalarization. Aiming at knowing the global structure of Pareto optimal set and/or Pareto frontier of a MOP with simple computation, we propose the basic concept that, (a) the SOPs with different weight should be solved independently by some evolutionary computation algorithms; and, (b) previously obtained useful solutions should be partly reused for the optimization of another SOP. In particular, for continuous-variable MOPs, we also propose a practical computation method based on the concept, which uses Particle Swarm Optimization (PSO) with effective reinitialization mechanism. The usefulness of the proposed methodology is demonstrated through numerical experiments with the proposed computation method.
    Download PDF (1875K)
<Information Processing, Software>
  • Takeru Yokoi
    2012 Volume 132 Issue 1 Pages 53-60
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    We propose here a novel approach to explore an optimized number of topics in a document set using consensus clustering based on Non-negative Matrix Factorization (NMF). It is useful to automatically decide the number of topics from a document set since various approaches to extract topics heuristically decide it. Consensus clustering merges multiple results of clustering so that it achieves a robust clustering. In this paper, assuming that a robust clustering is achieved by the optimized number of clusters, we have proposed a novel consensus soft clustering algorithm based on NMF and estimated an optimized number of topics with exploring a robust classification of documents into the topics.
    Download PDF (709K)
  • Shota Yunoki, Tomoki Hamagami, Kenji Oshige, Chihiro Kawakami, Noriyuk ...
    2012 Volume 132 Issue 1 Pages 61-67
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    A new call-triage system, a key part of emergency support system with stochastic network model is examined. The call-triage is an operation allowing the efficient decision of service grade and dispatching of suitable rescue team service from phone call information. Nowadays, the call-triage is being trialed on a few cities and is achieving an effect. However, there is the issue that if under-triage in which the condition of sick person is estimated more lightly is eliminated, the efficiency is degraded (over-triage). In this report, in order to overcome the issue, the Bayesian network scheme is examined to the call-triage system. The experiments with real call-triage data set results show the Bayesian network achieves precision enhancement.
    Download PDF (1198K)
Special Issue Letter
<Softcomputing, Learning>
Paper
<Electronic Materials and Devices>
  • Tsunemasa Saiki, Yuichi Utsumi
    2012 Volume 132 Issue 1 Pages 70-76
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Surface acoustic wave (SAW) devices can be used as component parts for micro total analysis systems (μTASs), because the structures of these are very simple. We thus fabricated a mixing chip with liquid flow actuators operated only by the SAWs. The chip is mainly consisted of one Y-type flow channel, two pumps and one mixer. The pump is located at each upstream area of merging point of the flow channel, and the mixer is located at its downstream area. Incidentally, the flow actuators have interdigital transducers (IDTs) to generate the SAWs. The IDTs were fabricated by patterning Al/Cr on a LiNbO3 substrate, and the flow channel wall was made of epoxy-based negative resist SU-8. Using the mixing chip, we carried out a mixing experiment of two water solutions with different concentrations by using glycerin, and a chemical reaction experiment of luminol luminescence. Through the experimentations, it was found that the local and rapid mixing of continuous liquid flow was possible by using the SAW actuators, and the availability of our fabricated mixing chip as a reactor was demonstrated.
    Download PDF (1259K)
  • Seiichiro Takahashi, Hironori Itoh, Ryuichi Komatsu
    2012 Volume 132 Issue 1 Pages 77-82
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Lifetime of an indium tin oxide (ITO) target is an important characteristic in the production of liquid crystal displays (LCDs). Increasing the sintering density of the ITO target is assumed to lead to an increased lifetime. So far, it has been clarified that the carbon concentration in In2O3 powder, the raw material of ITO targets, influences remarkably the target lifetime. In this study, with the aim of reducing the concentration of carbon in In2O3 powder, the synthesis of In2O3 powder containing dissolved Sn by a dry process was performed.
    Download PDF (1523K)
<Optics, Quantum Electronics>
  • Katsuhiro Mikami, Shinji Motokoshi, Masayuki Fujita, Takahisa Jitsuno
    2012 Volume 132 Issue 1 Pages 83-88
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    We evaluated the laser-induced damage resistance of dielectric coatings at temperature from 123 K to 473 K with 1064-nm wavelength at 4-ns pulse duration. In the results, laser-induced damage resistances were improved at low temperature for dielectric coatings (single layer of SiO2 and MgF2). These results are important for understanding the laser-induced damage mechanisms.
    Download PDF (712K)
  • Garuda Fujii, Toshiro Matsumoto, Toru Takahashi, Takayuki Yamada, Tsuy ...
    2012 Volume 132 Issue 1 Pages 89-95
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Random lasers are actively studied in recently years because of their unique properties which conventional laser devices do not have. In this research field, frequency responses of electromagnetic intensity in disordered media has been calculated conventionally by using FDTD method. However, recently by means of FEM, we presented some simulation results for laser action, namely, increases in the fluxes of Poynting vector. In the present study, by expressing the population inversion as a negative imaginary part of the dielectric constant, we evaluate both the Poynting vectors and electric intensities to assess the conventional studies, and investigate how the electric intensities in randomly distributed dielectric structure are related to laser actions. We discuss about the existence of laser modes caused by cavity mode and non-cavity mode and investigate laser mode distributions for population inversion density of optically active materials.
    Download PDF (1075K)
<Biomedical Engineering>
  • Takashi Numata, Yutaro Ogawa, Lui Yoshida, Kiyoshi Kotani, Yasuhiko Ji ...
    2012 Volume 132 Issue 1 Pages 96-103
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    The autonomic nervous system is important in maintaining homeostasis by mediating the opposing effects of the sympathetic and parasympathetic nervous activity on organs. Although it is known that the amplitude of RSA (Respiratory Sinus Arrhythmia) is an index of parasympathetic nervous activity, it is difficult to estimate that activity in real-time in everyday situations. It is partly caused by body motions and extrasystoles. Also, automatic recognition of the R-wave on electrocardiograms is required for real-time analysis of RSA amplitude, there is an unresolved problem of false recognition of the R-wave. In this paper, we propose a method to evaluate the amplitude of RSA accurately using statistical processing with probabilistic models. Then, we estimate parasympathetic nervous activity during body motion and isometric exercise to examine the validity of the method. As a result, using the proposed method, we demonstrate that the amplitude of RSA can be extracted with false recognition of the R-wave. In addition, an appropriate threshold for the estimate is one or five percent because waveforms of RSA amplitude do not follow the abrupt changes of the parasympathetic nervous activity evoked by isometric exercise with the threshold at ten percent. Furthermore, the method using normal distribution is found to be more appropriate than that of chi-square distribution for statistical processing. Therefore, we expect that the proposed method can evaluate parasympathetic nervous activity with high accuracy in everyday situations.
    Download PDF (686K)
  • Hiroaki Miyoshi, Takayuki Numata, Yutaka Kuwae, Masaki Sekine, Miwa Ts ...
    2012 Volume 132 Issue 1 Pages 104-110
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This study quantitatively compared lower limb motility of normal subjects and those requiring support level 1 (support_1). We developed a wireless inertia sensor with an embedded tri-axial accelerometer and angular velocity sensor. Six normal elderly subjects and ten elderly subjects who were classified as support_1 by the Japanese care insurance system participated in the study. We attached the wireless motion sensors to the center of the lower back and both thighs in the subjects. Subjects were then asked to walk 10 m and perform a stepping exercise. For the evaluation, the cadence, pitch angle, and pitch angular velocity of the thigh auto-correlation function and root mean square (RMS) on the lower back were calculated. The autocorrelation coefficient function for the support_1 subjects was smaller than in the normal subjects, while the RMS was larger in support_1. These differences indicated that the gait and balance abilities of the support_1 subjects were poorer than those of the normal subjects. This suggests that our wireless motion sensor is useful for assessing the motility of the lower limbs while walking and climbing steps.
    Download PDF (651K)
  • Jiaxing Ye, Takumi Kobayashi, Masahiro Murakawa, Tetsuya Higuchi, Nobu ...
    2012 Volume 132 Issue 1 Pages 111-119
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    In this paper, we propose an approach for abnormality detection in multi-channel ECG signals. This system serves as front end to detect the irregular sections in ECG signals, where symptoms may be observed. Thereby, the doctor can focus on only the detected suspected symptom sections, ignoring the disease-free parts. Hence the workload of the inspection by the doctors is significantly reduced and the diagnosis efficiency can be sharply improved. For extracting the predominant characteristics of multi-channel ECG signals, we propose multi-channel Fourier local auto-correlations (m-FLAC) features on multi-channel complex spectrograms. The method characterizes the amplitude and phase information as well as temporal dynamics of the multi-channel ECG signal. At the anomaly detection stage, we employ complex subspace method for statistically modeling the normal (healthy) ECG patterns as in one-class learning. Then, we investigate the input ECG signals by measuring its deviation distance to the trained subspace. The ECG sections with disordered spectral distributions can be effectively discerned based on such distance metric. To validate the proposed approach, we conducted experiments on ECG dataset. The experimental results demonstrated the effectiveness of the proposed approach including promising performance and high efficiency, compared to conventional methods.
    Download PDF (2040K)
  • Keiichiro Kagawa, Kenji Yamada, Eiji Tanaka, Jun Tanida
    2012 Volume 132 Issue 1 Pages 120-130
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    A focusing-free and multispectral three-dimensional endoscope based on a compact compound-eye camera is proposed. A prototype of TOMBO endoscope with 3×2 aspherical lenses using a wavefront coding technique and a 640×480-pixel color imager demonstrates extended depth of field and registration of multispectral images based on an estimated depth map.
    Download PDF (3327K)
<Systems, Instrument, Control>
  • Shinji Kitagami, Hitoshi Kamasaka, Yosuke Kaneko, Hisao Koizumi
    2012 Volume 132 Issue 1 Pages 131-140
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Remote monitoring and control services, in which industrial equipments are connected directly to the internet, have come into widespread use in recent years. The increase of remote services causes the service conflict problem: concurrent operation of multiple services on equipment causes interference and collision among these services. We propose a method of avoiding conflicts on remote services. In this method, service conflicts are avoided using rights of use which is delivered by owner of equipments to each service providers. The rights of use include acceptable commands and exclusive conditions. Comparing the conventional method which needs to distribute access control lists to equipments and needs to update them, the proposed method is easy to introduce into remote monitoring and control services. In this paper, we evaluate the proposed method using the simulation system where three remote services access an air conditioner. The results show that proposed method can avoid service conflicts without any complicated operations and with low cost of device implementation.
    Download PDF (759K)
  • Sangsoo Park, Yushi Miura, Toshifumi Ise
    2012 Volume 132 Issue 1 Pages 141-150
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This paper proposes an intelligent control for the distributed flexible network photovoltaic system using autonomous control and agent. The distributed flexible network photovoltaic system is composed of a secondary battery bank and a number of subsystems which have a solar array, a dc/dc converter and a load. The control mode of dc/dc converter can be selected based on local information by autonomous control. However, if only autonomous control using local information is applied, there are some problems associated with several cases such as voltage drop on long power lines. To overcome these problems, the authors propose introducing agents to improve control characteristics. The autonomous control with agents is called as intelligent control in this paper. The intelligent control scheme that employs the communication between agents is applied for the model system and proved with simulation using PSCAD/EMTDC.
    Download PDF (1110K)
  • Akiyoshi Ohno, Tatsushi Nishi, Masahiro Inuiguchi, Satoru Takahashi, K ...
    2012 Volume 132 Issue 1 Pages 151-159
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    In this paper, we propose a column generation for the train-set scheduling problem with regular maintenance constraints. The problem is to allocate the minimum train-set to the train operations required to operate a given train timetable. In the proposed method, a tight lower bound can be obtained from the continuous relaxation for Dantzig-Wolfe reformulation by column generation. The subproblem for the column generation is an elementary shortest path problem with resource constraints. A labeling algorithm is applied to solve the subproblem. In order to reduce the computational effort for solving subproblems, a new state space relaxation of the subproblem is developed in the labeling algorithm. An upper bound is computed by a heuristic algorithm. Computational results demonstrate the effectiveness of the proposed method.
    Download PDF (992K)
  • Hiroyuki Ono
    2012 Volume 132 Issue 1 Pages 160-165
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    This paper describes a high speed calibration method for laser positioner by scanning work area under constant velocity. Laser positioner consists of sinusoidal laser encoder, DC motor and the controller. The encoder has diffraction grating scale and laser optics. It generates sine and cosine outputs according to the travel distance. Qualities of the scale and the optics contribute to the fluctuation of output signal amplitude, DC offset and relative phase. First, the controller collects the output data under constant velocity of 4 samples per 1 sine wave length. The parameter can be estimated by simplified discrete Fourier transform method. Calibration data are collected every quarter sine wave length over all positioner work area. The result is stored in the table then referred by the controller in real time operation. Experiment results are also reported for HDD servo track writer application.
    Download PDF (651K)
<Softcomputing, Learning>
  • Yoko Akiyama, Fumihito Mishima, Shigehiro Nishijima
    2012 Volume 132 Issue 1 Pages 166-172
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    The objective of this study is to quantify the tactile sensation by physical indexes for material design of cosmetics or fabrics. We tried to indicate the sensory scores of tactile sensation as a combination of physical indexes. In order to extract the principal factors of tactile sensation, factor analysis was applied to the sensory evaluation data of skin lotions, whose tactile sensation is designed experimentally. Extracted factors were related to physical indexes by multiple regression analysis, and the sensory score was estimated by using the physical indexes such as viscosity, contact angle, friction coefficient and weight loss by temperature increase.
    Download PDF (549K)
Technical Note
<Infomation System, Electronic Commerce>
  • Yoji Yamato, Takaaki Moriya, Takeshi Ogawa, Junichi Akahani
    2012 Volume 132 Issue 1 Pages 173-180
    Published: January 01, 2012
    Released: January 01, 2012
    JOURNALS RESTRICTED ACCESS
    Recently, Cloud computing is spread rapidly and many Cloud providers start their Cloud services. One of the Cloud computing problems is Cloud provider Lock In” for users. Actually, Cloud computing management APIs such as ordering or provisioning are different in each Cloud provider, so that users need to study and implement new APIs when they change Cloud providers. OGF and DMTF start the discussions of standardization of Cloud computing APIs, but there is no standard now. In this technical note, to clarify what APIs cloud providers should provide, we study common APIs for Cloud computing. We survey and compare Cloud computing APIs such as Rackspace Cloud Server, Sun Cloud, GoGrid, ElasticHosts, Amazon EC2 and FlexiScale which are currently provided as public IaaS Cloud APIs in the market. From the survey, the common APIs should support REST access style and provide account management, virtual server management, storage management, network management and resource usage management capabilities. We also show an example of OSS to provide these common APIs compared to normal hosting services OSS.
    Download PDF (708K)
Letter
<Media Information, User Interface>
feedback
Top