Event‐related potentials affected by spatial frequencies of background visual pattern during a cognitive task

Spatial variation in luminance, which is one of the most important attributes of a visual image, may cause characteristic reactions in higher order brain functions. We aim to characterize the event‐related potentials (ERPs) associated with a cognitive task (simple addition) in terms of how they are influenced by the spatial frequencies of the background visual pattern. We use vertically striped visual stimuli with embedded numbers (0–9) and different spatial frequency characteristics (white noise, 1/f, 1/f², and 1/f³). The subjects are instructed to perform two tasks: an addition task that involved adding numbers presented as visual stimuli, and a reference task wherein the numbers were not added. In ERPs averaged over four types of visual stimuli for the addition task, positive components peaking at latencies of 182 ms in the central and frontal areas and 360 ms in the parietal area are observed, and significant differences are found between ERPs for the addition and reference tasks. In the addition task, the 182‐ms latency component shows a larger positive amplitude for 1/f³ compared with other stimuli in the right parietal‐occipital‐temporal area (P₄, T₆, O₂), and the 360‐ms latency component tends to show a larger positive amplitude for 1/f compared with other stimuli in the parietal‐central area (C₃, P₃, P_z). We conclude that spatial frequency characteristics influence ERPs associated with the retrieval of arithmetic data and certain neural activities that accompany simple forms of addition. © 2013 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Visual search controlled by spatial frequency without salient features in an image

To clarify the characteristics of visual search that are controlled by spatial frequency, we measured the scan path during eye movement in response to visual stimuli. A target stimulus with the spatial frequency of 1/f^β (where β is the slope of power spectral density of luminance in an image) was embedded in a background stimulus with white (1/f⁰) characteristics. An instruction‐based task was employed in this study, wherein a subject searches for the target as rapidly as possible and fixates on it. Participants were six healthy males aged 21–22 years. In the task, a threshold for spatial frequency that was the minimum value required by the subjects for the search corresponded to β =∼ 0.2 − 0.3. This finding has potential application in the design of visual presentation media, such as signs and tablets, by providing ease of viewing to the user. © 2015 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Utilizing gamma band to improve mental task based brain-computer interface design.

A common method for designing brain-computer Interface (BCI) is to use electroencephalogram (EEG) signals extracted during mental tasks. In these BCI designs, features from EEG such as power and asymmetry ratios from delta, theta, alpha, and beta bands have been used in classifying different mental tasks. In this paper, the performance of the mental task based BCI design is improved by using spectral power and asymmetry ratios from gamma (24-37 Hz) band in addition to the lower frequency bands. In the experimental study, EEG signals extracted during five mental tasks from four subjects were used. Elman neural network (ENN) trained by the resilient backpropagation algorithm was used to classify the power and asymmetry ratios from EEG into different combinations of two mental tasks. The results indicated that ((1) the classification performance and training time of the BCI design were improved through the use of additional gamma band features; (2) classification performances were nearly invariant to the number of ENN hidden units or feature extraction method.     

High‐tuning‐range CMOS band‐pass IF filter based on a low‐Q cascaded biquad optimization technique

A design procedure for high‐order continuous‐time intermediate‐frequency band‐pass filters based on the cascade of low‐Q biquadratic cells is presented. The approach is well suited for integrated‐circuit fabrication, as it takes into account the maximum capacitance spread dictated by the available technology and maximum acceptable sensitivity to component variations. A trade‐off between noise and maximum linear range is also met. A novel, wide‐tuning‐range transconductor topology is also described. Based on these results, a 10‐pole band‐pass filter for a code division multiple‐access satellite receiver has been designed and tested. The filter provides tunable center frequency (f₀) from 10 to 70 MHz and exhibits a 28‐MHz bandwidth around f₀ = 70 MHz with more than 39‐dB attenuation at f₀/2 and 2f₀. Third‐order harmonic rejection is higher than 60 dB for a 1‐Vpp 70‐MHz input, and equivalent output noise is lower than 1 mV_rms. The circuit is fabricated in a 0.25‐µm complementary metal oxide semiconductor process, and the core consumes 12 mA from a 2.5‐V supply, offering the best current/pole ratio figure. The die area resulted to be 0.9 × 1.1 mm². Copyright © 2014 John Wiley & Sons, Ltd.     

Man-machine communications through brain-wave processing

The possibility of monitoring voluntarily produced changes in the electroencephalogram (EEG) of a subject and translating these changes into a set of commands to be issued to an external device was investigated. Subjects performed five distinct tasks under both eyes-open and eyes-closed conditions. A feature set consisting of the asymmetry ratios and the power values for each lead at four frequency bands-delta (0-3 Hz) theta (4-7 Hz), alpha (8-13 Hz), and beta (14-20 Hz)-was used to characterize the EEG. The feature sets created from an estimate of the spectral density of the EEG for each task were used to test classification accuracy among the various tasks using a Bayes quadratic classifier. The results show that it is possible to distinguish, to a high degree of accuracy, among the various mental tasks studied, using only the EEG.     

Automatic EEG interpretation adaptable to individual electroencephalographer using artificial neural network

In view of the fact that the visual EEG interpretation could be a subjective task and may vary among the electroencephalographers, the main objective of this study was to develop an automatic EEG interpretation system which is adaptable to each electroencephalographer. The system adapted to each electroencephalographer would bring a close automatic EEG interpretation to that done by the electroencephalographer's visual interpretation. The adaptable automatic EEG interpretation was accomplished by using the constructive neural network with forgetting factor. The artificial neural network was constructed so as to give the integrative interpretation of the EEG based on the intermediate judgment of 13 items that characterized the visual interpretation. The developed method was evaluated based on the visually inspected EEG data of 37 patients by electroencephalographer‐A and the data of 20 patients by electroencephalographer‐B. The adapted ANN showed good agreement with each electroencephalographer's visual inspection. The proposed automatic EEG interpretation by use of the ANN can be a powerful assistant tool for individual electroencephalographers for their EEG interpretation. Copyright © 2001 John Wiley & Sons, Ltd.     

Harmonic fold back reduction at the N‐path filters

In this paper, a method is proposed to reduce harmonic fold back (HFB) problem of N‐path filters, without increasing the input reference clock (f_CLK ) frequency. The HFB at the N‐path filter is analyzed, and simple expressions are extracted to model this problem. Using the results of the analysis, an M‐of‐N‐path filter has been proposed that behaves like an M × N‐path filter in terms of HFB problem; however, the f_CLK frequency of this structure is the same as an N‐path filter. To demonstrate the feasibility of the proposed idea, a 3‐of‐4‐path filter is designed, and its characteristics are compared with 4‐path and 12‐path filters by simulation. Impacts of different non‐idealities like clock‐phase error, mismatch, and parasitic capacitance are investigated. The transistor‐level implementation of this filter is performed in 0.18 µm Complementary Metal Oxide Semiconductor (CMOS) technology. The simulation results show that the filter has the pass‐band gain of 17 dB, tuning range of 0.2–1.2 GHz, −3 dB bandwidth of 25 MHz, quality factor of 8–48, 18 dB out‐of‐band rejection, 16 dB rejection of the third harmonic of switching frequency (f_s ), and the noise figure of 4.35 dB (using ideal G_m cells) and 6.95 dB (for practical G_m cells). The strongest harmonic folding to the filter pass‐band occurs around 11f_s with the attenuation of 23.8 dB. Each G_m cell draws about 12.4 mA from 1.8 V supply, and the out‐of‐band IIP3 and P _{1 dB,CP} are 17 and 4 dBm, respectively. Copyright © 2016 John Wiley & Sons, Ltd.     

Calculation of frequency response of thyristor‐controlled series capacitor

In this paper, we propose a method of analytically calculating an impedance of a thyristor‐controlled series capacitor (TCSC). The TCSC consists of a thyristor‐controlled reactor (TCR) and a capacitor. If a small voltage component of frequency f superimposes the TCSC voltage of frequency f_o, then current components of frequency (nf_o ± f) flow through the TCR, where n takes even numbers. We derive theoretical equations for them. In the derivation, we consider an influence of odd harmonics in the TCSC voltage. We also consider deviations in its zero‐crossing times due to the superimposed voltage. The current components flow through the capacitor and the transmission line, and produce new voltage components of frequency (nf_o ± f). In order to count their influence, we introduce admittance matrices for the TCR, the capacitor, and the transmission line. By solving a network equation consisting of the matrices, we can obtain the distribution of the voltage and current components. We define the impedance of TCSC as a ratio of the voltage component of frequency f to the current one of the same frequency. The impedance analytically obtained corresponds well with that obtained by simulations. Its frequency responses vary with the firing method of thyristors. © 2002 Wiley Periodicals, Inc. Electr Eng Jpn, 139(3): 35–44, 2002; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.1158     

基于脑电自回归预测的实时相位估计方法

经颅电刺激(transcranial electric stimulation, TES)等无创刺激方式在与大脑内在神经电活动锁相时,能更有效的调节神 经振荡活动。 由于脑电(electroencephalogram, EEG)信号复杂的时变性,现有的方法难以同时满足相位估计精度和实时性的要 求。 为此,提出一种用于锁相刺激系统的实时相位估计方法。 该方法对 EEG 信号进行自回归(autoregressive, AR)建模,然后利 用 AR 模型预测 EEG 信号并进行相位特征点识别,再通过相位特征点计算出待刺激点的相位。 采用该方法对 20 名受试者(年 龄 20~ 36 岁,男性 12 名,女性 8 名)的闭眼静息 EEG 数据进行分析,发现该方法的性能与模型系数更新时长、预测步长及 EEG 的窄带功率大小有关,对高窄带功率的 EEG 数据具有更优性能;在最佳模型参数下(模型系数更新时长为 5 s、预测步长为 30), 20 名受试者的平均锁相指数(phase locking value,PLV)为 0. 968,平均相位误差(average phase error,APE)为 13. 33°。 相对于平 均周期法,该方法具有更高的 PLV 值和更低的相位误差,可用于闭环锁相经颅电刺激仪器的研发。     

Kinesthetic motor imagery modulates intermuscular coherence

Intermuscular coherence can identify oscillatory coupling between two electromyographic (EMG) signals, measuring common presynaptic drive to motor neurons. Beta band oscillations (15-30 Hz) are hypothesized to originate largely from primary motor cortex, and are reduced during dynamic relative to static motor tasks. It has yet to be established whether motor imagery modulates beta intermuscular coherence. Using visual feedback, 10 unimpaired participants completed eighteen trials of pinching their right thumb and index finger at a constant force. During the 60-second trials, participants simultaneously engaged in one of three types of kinesthetic imagery: the right thumb and index finger executing a constant force pinch (static), the fingers of the right hand sequentially flexing and extending (dynamic), and the right foot pushing down with constant force (foot). Motor imagery of a dynamic motor task resulted in significantly lower intermuscular beta coherence than imagery of a static motor pinch task, without any difference in task performance or root-mean-square EMG. Thus, motor imagery affects intermuscular coherence in the beta band, even while measures of task performance remain constant. This finding provides insight for incorporation of beta band intermuscular coherence in future motor rehabilitation schemes and brain computer interface design.     

基于多域信息融合的脑电情感识别研究

脑电信号识别方法较少将空间、时间和频率信息相融合,为了充分挖掘脑电信号包含的丰富信息,本文提出一种多域信息融合的脑电情感识别方法。该方法利用二维卷积神经网络和一维卷积神经网络相结合的并行卷积神经网络(PCNN)模型学习脑电信号的空间、时间和频率特征,来对人类情感状态进行分类。其中,2D-CNN用于挖掘相邻EEG通道间的空间和频率信息,1D-CNN用于挖掘EEG的时间和频率信息。最后,将两个并行卷积模块提取的信息融合进行情感识别。在数据集SEED上的情感三分类实验结果表明,融合空间、时间、频率特征的PCNN整体分类准确率达到了98.04%,与只提取空频信息的2D-CNN和提取时频信息的1D-CNN相比,准确率分别提高了1.97%和0.60%。并于最近的类似工作相比,本文提出的方法对于脑电情感分类具有一定的优越性。     

A new fast‐lock,low‐jitter,and all‐digital frequency synthesizer for DVB‐T receivers

Lock time and convergence time are the most important challenges in delay‐locked loops (DLLs). In this paper we cover French very high frequency band with a novel all‐digital fast‐lock DLL‐based frequency synthesizer. Because this new architecture uses a digital signal processing unit instead of using phase frequency detector, charge pump, and loop filter in conventional DLL, therefore, it shows better jitter performance, lock time, and convergence speed than previous related works. Optimization methods are used to make input and output signals of the proposed DLL in phase. The proposed architecture is designed to cover all channels of French very high frequency band by choosing number of delay cells in signal path. Simulation has been done for 22–27 delay cells, and f_REF = 16 MHz, which can produce output frequency in range of 176–216 MHz. Locking time is approximately 0.3 µs, which is equal to five clock cycles of reference clock. All of the simulation results show superiority of the proposed structure. Copyright © 2013 John Wiley & Sons, Ltd.     

Toward mapping spatiotemporal characteristics of muscle oxygenation in different motor modalities by multichannel near‐infrared spectroscopy

Commonly, the near‐infrared spectroscopy (NIRS) devices are used to measure muscle oxygenated hemoglobin (HBO) concentration and total hemoglobin (HBT) concentration with a single channel, which can obtain the temporal HBO and HBT signals. Lacking of the spatial information of muscle oxygenation will limit the exploration of the heterogeneity of muscle activities. In this study, a multichannel NIRS recording system was used to measure the muscle oxygenation with an attempt to simultaneously provide the temporal and spatial HBO and HBT concentration. In the experiment, the influences of four motor tasks, including active, passive, imaginary movement, and the control task (no movement), on muscle oxygenation were investigated in eight normal subjects. Our results showed that both amplitude and spatial heterogeneity associated with variation in muscle oxygenation during active and passive motor tasks were significantly different between the motor time and rest time (P < 0.05). Furthermore, the region where HBO concentration decreased during the active motor task was in accordance with the anatomical position of contracted muscle, which cannot be observed from the results of the passive task. Considering the imaginary and control tasks in which the muscle was in a fixed/stationary state, the amplitude and spatial heterogeneity associated with variations in muscle oxygenation only exhibited slight changes (P > 0.05). This pilot study suggested that the spatiotemporal information obtained from multichannel NIRS devices might be potential for accurate measurement of the variation in muscle oxygenation during motor tasks, which would be useful for different clinical applications.     

Multi‐component instantaneous frequency estimation using locally adaptive directional time frequency distributions

This paper presents a locally adaptive time‐frequency (t,f) method for estimating the instantaneous frequency (IF) of multi‐component signals. A high‐resolution adaptive directional time‐frequency distribution (ADTFD) is defined by locally adapting the direction of its smoothing kernel at each (t,f) point based on the direction of the energy distribution in the (t,f) domain. The IF of signal components is then estimated from the ADTFD using an image processing algorithm. Using the mean square error between the original IF and estimated IF as a performance criterion, experimental results indicate that the ADTFD gives better IF estimation performance compared with other TFDs for a multi‐component signal. For example, for signal‐to‐noise ratio of 12dB, the IF estimate obtained using the ADTFD achieves a mean square error of ?42dB for a weak signal component, which is an improvement of ?12dB compared with other TFDs. Copyright © 2015 John Wiley & Sons, Ltd.     

An example of dB/dt characteristics of the dynamic magnetic loss parameter

Ferrite cores are used for power electronics circuits operating under excitation conditions of frequency ranging from hundreds of kilohertz to several tens of megahertz. Development of a mathematical model enabling the estimation of the power loss of ferrites in high‐frequency regions, contributes to the designing of power circuits. The power loss in high‐frequency regions consists of hysteresis and dynamic magnetic losses. Although hysteresis loss does not depend on the excitation frequency, the dynamic magnetic loss increases with increasing excitation frequency. The dynamic magnetic loss parameter, λ_f, has been introduced for identifying the power loss in high‐frequency regions. In this study, λ_f − dB/dt characteristics are experimentally obtained by a newly introduced hysteresis model that uses a normal distribution function. It is found that the λ_f − dB/dt characteristics do not depend on the size of B − H loops. The instantaneous power loss of a ferrite core in the high‐frequency region can be calculated using the λ_f − dB/dt characteristics. Copyright © 2007 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Periodic noise analysis of electric circuits: Artifacts,singularities and a numerical method

In this paper it is shown that a numerical method largely adopted for the simulation of noise in autonomous circuits is affected by singularities that manifest when the frequency at which the noise analysis is carried out approaches a harmonic of the autonomous circuit. The resulting noise power spectral density (PSD) is thus characterized by spurious spikes. The presence of these singularities is for the first time justified from an analytical standpoint and their effects are shown by simulating some oscillators, employed as benchmarks. Furthermore, the presented approach justifies the 1/(f_s?f)² shape of the PSD of noise at the output when the f_s frequency approaches the f fundamental of a stable oscillator and the 1/|f_s?f|³ shape when the effects of flicker noise are manifest. Copyright © 2009 John Wiley & Sons, Ltd.     

Cognitive tasks for driving a brain-computer interfacing system: a pilot study

Different cognitive tasks were investigated for use with a brain-computer interface (BCI). The main aim was to evaluate which two of several candidate tasks lead to patterns of electroencephalographic (EEG) activity that could be differentiated most reliably and, therefore, produce the highest communication rate. An optimal signal processing method was also sought to enhance differentiation of EEG profiles across tasks. In ten normal subjects (five male), aged 29-54 years, EEG activity was recorded from four channels during cognitive tasks grouped in pairs, and performed alternately. Four imagery tasks were: spatial navigation around a familiar environment; auditory imagery of a familiar tune; and right and left motor imagery of opening and closing the hand. Signal processing methodology included autoregressive (AR) modeling and classification based on logistic regression and a nonlinear generative classifier. The highest communication rate was found using the navigation and auditory imagery tasks. In terms of classification performance and, hence, possible communication rate, these results were significantly better (p < 0.05) than those obtained with the classical pairing of motor tasks involving imaginary movements of the left and right hands. In terms of EEG data analysis, a nonlinear classification model provided more robust results than a linear model (p < 0.01), and a lower AR model order than those used in previous work was found to be effective. These findings have implications for establishing appropriate methods to operate BCI systems, particularly for disabled people who may experience difficulty with motor tasks, even motor imagery.     

Using residue sums to estimate high‐order Fourier harmonics of piecewise‐continuous transcendental functions: Application to Class A‐, B‐ and F‐type amplifier circuits

A discrete fast‐Fourier transform (DFFT) is the preferred method of choice for the rapid evaluation of a set of harmonics of a piecewise‐continuous and periodic transcendental form. For large sets of Fourier components specified to some stringent error criterion, the approach becomes increasingly unattractive owing to the presence of round‐off errors that result from the switching of one transcendental form to another. As an alternative, it might be wondered whether the high‐frequency components can be more efficiently estimated by employing a combination of residue sums and boundary integrals in the complex plane z = Re^iωt, where ω is the fundamental frequency and R = ∣z∣. The starting point is the construction of suitable contours that divide the complex plane into a number of sectors in accordance with the number of intervals of smooth behaviour of a periodic piecewise‐continuous real function along ∣z∣ = 1. Each sector encompasses the analytic extension of a real transcendental function on ∣z∣ = 1 to yield p(z)T(f(z)), where T(ζ) is meromorphic and p(z), f(z) are Laurent series. Fourier coefficients are subsequently expressed in terms of residue series and constant‐phase boundary integrals from each of the various sectors associated with a given p(z)T(f(z)). This approach is applied to the model for the drain current of a field effect transistor (FET), where in this case T(ζ) = tanh(ζ), which is subject to the modes of operation: ‘Class A’, ‘Class B’ and approximate ‘Class F’. In contrast to Classes A and B, the Fourier coefficients in the ‘Class F’ drain current decay slowly with frequency, suggesting that this mode might be more suitably analysed using a combined DFFT/residue procedure. Copyright © 2007 John Wiley & Sons, Ltd.     

Insulation characteristics of GIS under nonstandard lightning impulse oscillations: Insulation characteristics under double‐frequency oscillations with various frequencies and damping ratios

To improve GIS insulation specifications, it is important to recognize the insulation characteristics under oscillatory overvoltage waveforms occurring in the field. This paper describes investigations of insulation characteristics for double‐frequency oscillatory waveforms with various frequencies and damping ratios. It was found that minimum breakdown voltages (V_min) were irrelevant to frequency changes in the range of experimental conditions under the same damping conditions. On the other hand, V_min rose with the damping ratio rising under the same frequency conditions. The insulation characteristics at a valley of oscillation were investigated using actual breakdown voltages. It was found that the insulation characteristics were treated all‐inclusively based on the characteristics of V_min for damping time. © 2003 Wiley Periodicals, Inc. Electr Eng Jpn, 146(2): 11–19, 2004; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.10245