Electronics and Communications in Japan最新文献

Assessment of impaired insulin secretion and insulin resistance is critical to determining the type of diabetes and planning optimal management and preventive strategies in clinical practice. To measure insulin resistance, a minimally invasive needle with measurement and administration flow channels for local glucose tolerance tests was fabricated. The administration flow channel has a hole on the tip end of the flow channel. The microperfusion flow channel starts from the root side of the needle, turns around at the needle-tip end, and returns back to the root side of the needle. It has a perforated membrane on the surface of the area placed in the skin when inserted. Glucose is administrated through the flow channel, and glucose concentration of the skin is measured by microperfusion before and after administration. When the perfusate is perfused into the microperfusion flow channel, since the glucose concentration of the skin is higher than the perfusate, glucose contained in the skin's interstitial fluid moves into the flow channel through the perforated membrane by concentration diffusion, and the glucose can be collected. The needle's administration and glucose-collecting functions were evaluated in the skin of mice.

Facial Expression Recognition has been studied for many years; however, it remains a challenging task in real-world environments due to complex backgrounds, varying illumination conditions, and online processing issues. In this study, we propose a deep learning model, CAER-Net-RS, by leveraging multiple training datasets. The proposed model integrates three neural networks: the Face Network, the Context Network, and the Adaptive Network. Different datasets are employed for the pretraining of these networks: the facial expression image dataset RAF-DB for the Face Network, the scene image dataset Places365-Standard for the Context Network, and the CAER-S dataset for the Adaptive Network. In the experiment, the proposed model achieved an average recognition accuracy of 85.20% across seven types of facial expressions, compared to 70.92% for the conventional Context-Aware Emotion Recognition Network (CAER-Net).

Previously, the authors have proposed a contactless probe-type current sensor with low insertion impedance. In this paper, the insertion impedance of the proposed current sensor at high frequencies is clarified. Then, we measured the actual high-speed switching current flowing through a GaN device and compared the measured waveforms with those of a coaxial-type shunt resistor. Comparison of the measured waveforms showed that the coaxial shunt resistor exhibited ringing in the waveform due to the effect of insertion impedance, but the proposed current sensor exhibited less ringing, indicating that the measurement is minimally invasive to the circuit under test.

To improve the accuracy of the sound image localization technique using bone conduction headphones and head-related transfer functions, this paper proposes a preprocessing method using the difference of transfer characteristics between from the entrance of ear canal to cochlea and from transducer to cochlea. The estimation of the differential characteristic is performed using Distortion Product OtoAcoustic Emission (DPOAE) and the preprocessing is realized using finite impulse response (FIR) digital filters. In subjective experiments on sound image localization, it was confirmed that the proposed method improves the localization accuracy compared to the case without preprocessing.

In control systems using network communication or digital sensors, the quantization of input and output signals can significantly impact control performance. Therefore, it is essential to incorporate signal quantification into the control system design to prevent performance degradation. This study proposes a controller design method for systems in which both inputs and outputs are quantized. The proposed method derives a controller that minimizes the upper bound of the error between a plant output and a reference model output. In contrast to previous studies, our proposed method avoids solving matrix inequalities numerically and provides an analytical design approach.

An antenna beamforming system with feedback phase control is proposed, and an analysis of the control stability is described. When the number of the antenna elements is two, the phases are controlled depending on the residual phase difference of the signals. It can be realized with a simple feedback system. When the number of the antenna elements is more than two, there are plural phase differences, and plural control loops are required. It makes the stability analysis complex. In this paper, a control system with small number of feedback loops is proposed, and an analysis method with eigenvalues of the transfer matrix is described.

We developed a static magnetic field for miniaturization of the nuclear magnetic resonance (NMR) measurement system to measure nutrient solution in agricultural field. Instead of the large electromagnets used in our previous work, a permanent magnet was used to establish a static magnetic field. The design was based on the Halbach array, which allows the magnetic field to be concentrated on one side to increase the magnetic force. To avoid the problem of reducing the magnetic force of the magnets, the array magnets were optimized so that they could be composed of a combination of simple square prism magnets. Observation of the magnetic flux density of the fabricated permanent magnets showed that the magnetic flux density at the center was 0.946 T, and an error of less than 5 mT in magnetic flux density could be achieved. As a result, we succeeded in producing the permanent magnet. NMR measurements of water and air were performed using the produced magnets. When measuring water, a signal at 39.679 MHz was observed, which is the signal of hydrogen. In contrast, no signal was observed in air. Therefore, the NMR signal was successfully measured using the fabricated permanent magnet.

There is a lot of previous research on cultivation management based on mathematical engineering. In some previous research, the expected profit of cost-constrained cultivation management for multiple agricultural fields with sensors is maximized under the condition that growing states are unknown. However, in the previous research, it is assumed that the same crop will be grown in all agricultural fields and crop prices do not fluctuate. In this research, cost-constrained cultivation management with sensors considering fluctuation of crop price under the condition that different crops can be grown in the agricultural fields and growing states are unknown is studied. The cultivation management problem is modeled by Markov decision processes, as in previous research. The expected profit is maximized with reference to a Bayes criterion using dynamic programming in the proposed method. The effectiveness of the proposed method is shown by some computational examples. Adaptive cost adjustment between the different crops is confirmed in the computational examples.

This paper presents a CSI-based human detection system with commodity sensors: M5Stack. Especially, we remedy too sensitivity to individual with respect to measured CSI. Monitoring systems considering privacy proof is important for monitoring the elderly and children. In this paper, we carry out experiments on indoor human location estimation using Channel State Information (CSI). Specifically, to improve recognition accuracy in cases where training data and test data are generated from different persons, we proposed a method that includes dimensionality reduction using Principal Component Analysis (PAC) as a preprocessing. In our experiments, we clarify problems of CSI-based method and solve the problems using the proposed method. In the evaluation experiments, we achieved an improvement of 10%–30% depending on individuals.

Automatic music transcription (AMT) is a task that automatically generates music scores from the sound of music. Pitch estimation is one of the most important technologies in AMT, and its accuracy strongly affects the quality of AMT. Although its accuracy is improving, thanks to deep neural networks nowadays, pitch estimation for multi-instrumental music is still a challenging task, because there are many difficulties caused by numerous and complicated timbre patterns in multi-instrumental music. In this paper, we propose a pitch estimation model based on the reconstruction of the spectrum of input sound which can reduce the effect of timbre complexity on pitch estimation. The experimental result indicated that the proposed model improved the average precision by 8.8 points compared to the conventional method.