電気学会論文誌C(電子・情報・システム部門誌)
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
138 巻, 12 号
選択された号の論文の25件中1~25を表示しています
特集:電気・電子・情報関係学会東海支部連合大会
特集論文
<電気回路・電子回路>
  • 野崎 佑典, 吉川 雅弥
    2018 年 138 巻 12 号 p. 1455-1463
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    The illegal attacks for embedded devices have recently been reported. Therefore, lightweight block ciphers, which are suitable for embedded devices, have been attracted attention. Midori is a lightweight low power block cipher. Regarding the hardware security, the risk of power analysis for the cryptographic hardware is pointed out. Hence, the tamper resistance evaluation of a lightweight low power cipher Midori is very important to guarantee the security of embedded devices. However, power analysis of Midori has not been studied. Therefore, this study proposes a new power analysis method for Midori. The proposed method performs the hierarchical power analysis to analyze the all secret keys. Moreover, to reduce the computational complexity, the proposed method performs the efficient improvement method utilizing the characteristic of Midori algorithm. Experiments using a field programmable gate array prove the validity of the proposed method.

<情報通信工学>
  • 田渕 茉也, 伊藤 嘉浩
    2018 年 138 巻 12 号 p. 1464-1468
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    This paper proposes a method of estimating the most appropriate transmission rate of an IEEE 802.11g wireless LAN based on QoE for Web services. The method utilizes the linear multiple regression analysis to find a formula for calculating QoE from QoS; it then estimates the most appropriate transmission rate by using the multiple regression line. This paper confirms the effectiveness of the method by experiment. In the experiment, the authors target actual two Web services, a shopping service and a map service; they consider the round trip time and the packet loss ratio as QoS parameters and treat the satisfaction as a QoE parameter. The authors performed QoS evaluation and QoE assessment. The experimental results show the appropriateness of the method.

<システム・計測・制御>
  • 岡村 雅行, 弓場井 一裕, 矢代 大祐, 駒田 諭
    2018 年 138 巻 12 号 p. 1469-1477
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Generally, there are unmodelled dynamics and uncertainty between an actual plant and its mathematical model. which causes performance degradation or destabilization of the control system at worst. This paper focuses on mini-mization of performance degradation between the nominal performance and the actual performance by appropriately designing a Youla parameter in the presence of unmodelled dynamics and uncertainty of the plant. By introducing the Youla parameter and the dual Youla parameter, control performance is analyzed using coprime factorizations of the plant and the controller, and then a design method of a fully-parameterized fixed-order Youla parameter based on frequency responses of the plant is formulated for Single-Input-Single-Output systems as a convex optimization problem. The one of merits of the proposed method is to directly design a low-order Youla parameter from frequency responses without any controller reduction techniques. The effectiveness of the proposed method is confirmed by some experiments.

  • 権田 貴紀, 元谷 卓, 道木 加絵, 鳥井 昭宏
    2018 年 138 巻 12 号 p. 1478-1484
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    In this study, the effect of air space in waterproof sealed case of undersea magnetic field resonance type wireless power transfer (WPT) is reported. It has been reported that the transmission power is increased by the air space between the transmitter and receiver coil in the waterproofing case and the wall surface of the case. We evaluated the dependence of the thickness of the air space from measurement and magnetic field analysis. The measurement and magnetic field analysis results indicated the similar tendency to increase the transmission power by 9.8% with the 10 mm air space at the transmitter side and the 30 mm air space at the receiver side.

<メディア情報,ユーザ・インタフェース>
<ソフトコンピューティング・学習>
  • 内藤 昂佑, 加藤 昇平
    2018 年 138 巻 12 号 p. 1493-1499
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    In recent years, Social Networking Serveices (SNSs) have been popular among young people. According to the popularity of SNSs, the cyberbullying becomes a social problem. To find the mechanism of cyberbullying, there are several studies using multi-agent simulation which can analyze human relationships. Therefore, in this study, we model SNSs in classroom and analyze the influence of SNSs on classroom friendship. This model is based on the Socion theory. In the model, an agent can communicate other agents using two kinds of networks: a network of classroom and networks of SNSs via a network recognized by each individual. Agents communicate face to face (FTF) in the classroom and do SNS communication in SNSs. In this paper, we conduct simulation of friendship relations considering SNSs, and disccuss the influence of SNSs on classroom relationships from the simulaion results. As a result, the average favorability rating of agents increased through SNSs. On the other hand, a result that specific agent was rejected also appeared. We consider that information sharing by SNSs is related to increased bullying.

<情報処理・ソフトウェア>
  • 河村 洋子, 田島 孝治, 山田 博文, 所 哲郎
    2018 年 138 巻 12 号 p. 1500-1507
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    National Institute of Technology (NIT), Gifu College applied for the “Acceleration Program for University Education Rebuilding” started by the Ministry of Education, Culture, Sports, Science and Technology in 2014, and our application, a composite type of project (Theme I & II) was adopted. As one of our activities, we built “a collection of subjects highly recommended by company engineers” together with senior graduates of NIT, Gifu College. Then, we provide the subjects to the student as the CBT (Computer Based Testing) subjects using Moodle. Furthermore, we introduced a learning support system where students can gain the “practical engineering credit point” through taking the subjects. Through the learning support, the number of the LMS access to “a collection of subjects highly recommended by company engineers” increased by 1.3 times from last year.

  • 佐藤 秀樹, 中山 泰宏
    2018 年 138 巻 12 号 p. 1508-1516
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Geographical information query <D, F, ψskyline> retrieves a subset from geographical objects D under multi-criteria, each of which is an aggregate network distance in road networks and expressed by function fi(p, Qi) (∈ F) where p is an geographical object and Qi is a set of query points. ψskyline is skyline operation which is used to deal with multi-criteria. To process geographical information queries, Basic algorithms have been developed before. Two techniques, (1) shortest path search using path information cache and (2) incremental computation of aggregate network distances, are presented to improve processing performance of Basic algorithms. By applying both techniques to BasicLBC algorithm which is the most efficient among Basic algorithms, 2 algorithms, IncrementalLocalLBC and IncrementalGlobalLBC, are developed. Experimental results show that processing time of IncrementalGlobalLBC is reduced to 18% of that of BasicLBC, which is equal to 3% of that of BasicCE, an naive Basic algorithm.

論文
<電気回路・電子回路>
<情報通信工学>
  • 佐々木 範雄, 陳 強, 安達 文幸
    2018 年 138 巻 12 号 p. 1529-1538
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    In this paper, we study a minimum mean square error based frequency-domain equalization (MMSE-FDE)scheme for a high-speed digital power line carrier system achieving 1.1 Mbps using 300 kHz bandwidth. Training sequence (TS) inserted single-carrier (SC) block transmission is employed. In the TS-SC block transmission, the Zadoff-Chu sequence is used as TS. TS acts as the cyclic prefix (CP) for MMSE-FDE. The channel estimation and noise power estimation are done using TS for computing the MMSE-FDE weight. In this paper, we design a digital power line carrier system of 210 ksymbol/s, in which the TS length of 64 symbols (Zadoff-Chu sequence) and the data block length of 960 symbols (64QAM) are used for achieving 1.1 Mbps transmission. The transmission performance of the designed digital power line carrier system is evaluated by computer simulation in terms of the normalization mean square error (NMSE) of channel estimation, the mean square error (MSE) of MMSE-FDE, and the bit error rate (BER). It is confirmed that MMSE-FDE is superior to MMSE time-domain equalization (MMSE-TDE) and zero-forcing (ZF) FDE and it achieves the required BER of 1×10-6 when the received Eb/N0=25 dB. Furthermore, we clarified the optimum value of forgetting factors of the first order infinite impulse response (IIR) filters used for channel estimation and noise estimation.

<生体医工学・福祉工学>
  • 三上 可菜子, 大浦 邦彦
    2018 年 138 巻 12 号 p. 1539-1546
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    This paper discusses dynamics of oxygenated hemoglobin concentration and electrical signal that comes from inner brain activity relating to visual attention. Attention is well known as a base of cognitive function in neuroscience. Attentional disorder appears frequently in higher order brain dysfunction. In our previous studies, paced auditory serial addition test (PASAT) which is one of the examinations in neuropsychology, has been used to examine auditory attention and measure hemodynamics on prefrontal cortex. The authors changed their target to visual stimulus and proposed paced visual serial addition test (PVSAT) for experiment. Considering the results of hemodynamics or assistive event related electrical potentials, some interesting suggestions are given.

<システム・計測・制御>
  • 中川 善継, 村上 知里, 森 一幸, 佐藤 春彦
    2018 年 138 巻 12 号 p. 1547-1553
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    In radiation measurement after decontamination in nuclear disaster areas affected by nuclear power plants, it is known that the local high dose sites called hot spots are scattered. The measurement method of efficient and simple and also different from conventional method is required in order to search for local hot spot factors. This paper describes the estimation method of using model possessed the nonlinearity as the way to minimize the influence of both mechanical and stochastic noise included in ether the sensors or data, regarding time-series data monitored by using external sensors. This method could convert to the amount of radiation detection supposed from relative value of it in a static detection at the point by paying attention of change of peak in radiation detection level according to moving velocity in identical amount of radiation detection. We presume the parameters of the model with repeating the least squares method computation steps consisting of a number of finite sampling transition from the starting point of searching in time series data based on a function model according to moving velocity while moving multiple times, and derive the relation to dose level converted to radiation detection indicating for the behavior of hot spot factors at the point.

  • 武田 健吾, 澤田 賢治, 横川 慎二, 新 誠一
    2018 年 138 巻 12 号 p. 1554-1565
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Wind and solar power generators cause electric power fluctuation, batteries are combined with such the renewable energy system to absorb power fluctuations, so called, wind turbine and photovoltaic power generators and battery hybrid system (WPBHS). Such the system mainly focuses on the decentralized operation in which a large number of system devices are managed decentrally. On the other hand, its stable generating capability and fault tolerance performance depends on how to group devices into small groups in which components are managed centrally (We call it grouping). That is, the grouping itself can improve the decentralized operation performance. This paper considers overlapping grouping of WPBHS in which one device belongs to multiple groups and proposes a new decentralized operation. To verify the validity of the proposed method, this paper shows numerical experiments in terms of stable generation and fault tolerance.

  • 村田 頼信, 西川 美沙都, 松田 正誉
    2018 年 138 巻 12 号 p. 1566-1572
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    In the process of manufacture of silicon wafers, it becomes a problem that microscopic defect called microcrack may occur in a wafer. It is difficult to detect by an image inspection using infrared or a usual ultrasonic inspection because the aperture of microcrack is closed. In this study, as a nondestructive testing for a silicon wafer, we paid attention to inspection method using the nonlinearity of ultrasonic wave. First, a system which detects microcracks by making ultrasonic wave propagate along the surface of a silicon wafer was constructed. However, the ultrasonic wave propagating in a silicon wafer was not identified, and the optimization of the system had been unsolved. Therefore, the propagation analysis of the ultrasonic wave was carried out. As the result, it was suggested that it is a Lamb wave. Finally, by using the characteristic of Lamb wave, it was demonstrated that the proposed inspection method was useful for not only the microcrack inspection which lengthened the propagation distance but also the microcrack inspection in the back surface.

  • 大津山 卓哉, 本田 純一, 渡邊 優人
    2018 年 138 巻 12 号 p. 1573-1579
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Recently, study of Multi-Static Primary Surveillance Radar (MSPSR) which is expected to be used as the next generation primary surveillance radar (PSR) has been active in the filed of aircraft surveillance. MSPSR is one of passive radar which combine multiple receivers and/or transmitters, and many research and development projects are progressing with various radio system as transmitter source.

    Electronic Navigation Research Institute (ENRI) developed an optical fiber connected passive primary surveillance radar (OFC-PPSR) for the effective use of frequency band and the current PSR system enhancement. In this paper, we present the introduction of OFC-PPSR and the preliminary evaluation results of the OFC-PPSR system.

  • 岡田 将吾, 増田 士朗
    2018 年 138 巻 12 号 p. 1580-1585
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    The paper proposes data-driven minimum variance regulatory control using a Fictitious Reference Iterative Tuning (FRIT) method. Our previous work has extended the original FRIT for reference signal tracking to the case of a stochastic disturbance attenuation problem. However, the method parametrizes the controller by using plant parameters based on a specific controller structure for minimum variance control. The present work uses a linearly parametrized controller for the inverse of the controller. Thus, a general parametrization is realized, and the integrator is straightforwardly introduced. An analytical result for the data-driven cost function is also given. A numerical example shows effectiveness of the proposed method.

<知能,ロボティクス>
  • 多田 拡太郎, 杉浦 彰彦
    2018 年 138 巻 12 号 p. 1586-1594
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    This study examines a method of applying a multistep CNN (Convolutional Neural Network) by creating the surface shape images of the inner surface of a tire to classify defects occurring on a tire inner surface. A multistep classifier is made by using the symmetry of the ratio of the number of the training images for each classification label and the symmetry of the characteristics for each classification label. The examined method shows 17.9% improvement in the defect detection rate and 0.009% reduction of the area rate of overdetection regions by using symmetry, reducing overdetection while maintaining the accuracy of detecting defects. Also, in the environmental condition of a working state, it shows the defect detection rate of 89.3% and the area rate of overdetection regions of 0.049%.

<音声画像処理・認識>
  • 植田 祥明, 古賀 崇了, 末竹 規哲, 内野 英治
    2018 年 138 巻 12 号 p. 1595-1603
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Steganography is a technique to hide a secret data into stego-data for secret communications. In the steganography, it is important to suppress a degradation of the stego-image for data hiding. The stego-image of the steganographic method is divided into full-color images and palette-based images. In this paper, we propose a new multi-bit steganographic method for the palette-based images. In the case of embedding for palette-based images, there are two characteristics described below. (1) Embedding degradation is not remarkable because palette-based images have been degraded due to color quantization and error-diffusion processes. (2) The embedding error tends to greater than full-color images due to usable colors are few. In the proposed method, representative colors of the palette-based image are sorted and their index numbers are replaced so that the embedding error becomes small. The proposed method suppresses the degradation of the cover-image by dispersing the embedding coordinates. We verify the effectiveness of our method through experiments using palette-based color images.

  • 中村 悦郎, 髙橋 毅, 景山 陽一, 石井 雅樹, 西田 眞
    2018 年 138 巻 12 号 p. 1604-1612
    発行日: 2018/12/01
    公開日: 2018/12/01
    ジャーナル フリー

    Lip motion features such as changes in the lip width and length provide important information for analyzing psychology and physical conditions. Furthermore, extracting lip motions from facial images of users has some advantages that are "needless of special equipment" and "possible of non-contact measuring". Our previous studies revealed that lip motions relate to mental state. Therefore, it is important to develop a method for extracting the lip shape from facial image data. However, the conventional method failed to extract lips due to individual differences of each subject and influences of the shadows on the periphery of lips. Lip extraction methods based on feature points need to use a lot of learning data. In this paper, we propose a novel lip extraction method using feedforward neural network (FFNN), which is one of machine learning methods. The proposed method is able to learn the features of lips using only one image, and its accuracy is same or higher than that of the conventional method and feature-point-based methods.

<情報処理・ソフトウェア>
研究開発レター
<ソフトコンピューティング・学習>
部門記事
 
feedback
Top