The sound of the tube amp due to the harmonic distortion has been thought to depend not only on the vacuum tube type but also on the vacuum tube manufacturer and individual differences. However, quantitative differences in tube characteristics of the manufacturer have yet to be revealed. In this study, the characteristics of the triode 12 AX 7 of the five major manufacturers were analyzed based on statistical analysis of advanced physical model parameter sets. Tube characteristics were divided into three groups by cosine similarity and group clustering method. As a result, we succeeded for the first time quantitatively combining tube distortion characteristics and tube characteristics.

This paper presents input bias current (Ibias) reduction technique for high impedance CMOS op-amps with the proposed current compensation circuit to deal with the leakage current caused by Electro-Static Discharge (ESD) protection circuit of the IC. High input impedance CMOS op-amps are widely used for the application of high precision sensors with quite small input current. However, the leakage current of ESD protection circuit for op-amp causes a non-ideality error of the Ibias. Especially, the ESD leakage current increases drastically at the high temperature environment, and hence the Ibias of CMOS op-amp also increased significantly. An ESD leakage current compensation circuit is introduced to reduce the Ibias of CMOS op-amp. The prototype amplifier with the proposed current compensation circuit is designed and fabricated in standard 0.7 µm CMOS technology. Measurement results show that the Ibias is reduced to a 100 pA or less from a typical 2.3 nA at 150°C.

In recent years, interest in IoT (Internet of Things) has been increasing, and energy harvesting has attracted attention. However, the power and voltage that can be harvested are very small. Therefore, a power supply circuit is required to solve this problem. In this paper, we have realized a boost circuit that does not require a coil with the aim of integrating all the circuits required for sensing on one chip.

An exponentiation conversion circuit, which can be changed its power exponent to any value to compensate nonlinearity in electronic devices, is proposed. In the proposed circuit, new exponential conversion circuit converts signal multiplied logarithmically transformed input signal by the power exponent value, thereby obtaining the exponential power raised power function characteristic. The proposed circuit is a small scale circuit by utilizing an exponential characteristic in the subthreshold operation of MOSFET. For example, proposed circuit is suitable for integrating on a microcomputer chip used for IoT. The performance of the circuit was evaluated by prototype IC made by 0.6 µm CMOS process. In circuit simulation of the prototype IC, the exponent could be varied in the range of 0.038 to 2.46. By proposing the cascode exponential conversion circuit, the signal dynamic range was expanded.

In this paper, an auto-zero amplifier which reduces the offset voltage and output voltage ripple caused by the non-ideal characteristics of CMOS switches is proposed. Three capacitors are used to store the compensation voltage for an internal amplifier in the proposed circuit. The capacitors hold the voltage between two input terminals of the amplifiers even if a charge injection and/or a clock feed through occur at CMOS switches. The proposed circuit reduces the output offset voltage to 43.8% and the output voltage ripple to 42.8% compared to the conventional circuit.

In this paper, a third-order complex RⁱCR filter using small number of grounded imaginary resistors is proposed. The proposed complex RⁱCR filter can be obtained by adopting the lowpass-highpass transformation and an equivalent transformation for the conventional complex RⁱCR filter including a floating type imaginary resistor. The validity of the proposed method is confirmed through experiment.

This paper proposes a fast high-quality three-dimensional (3D) compressed sensing for a phased array weather radar (PAWR), which is capable of spatially and temporally high-resolution observation of the atmosphere. Because of the high-resolution, the PAWR generates huge observation data of approximately 500 megabytes every thirty seconds. To transfer this huge data in a public internet line for real time weather forecast, an efficient data compression technology is required. The proposed method compresses the PAWR data by randomly transferring several measurements only in the troposphere, and then reconstructs the missing measurements for each small 3D tensor data by minimizing a cost function based on a prior knowledge on weather phenomena. The minimizer of the cost function can be quickly computed by using a convex optimization algorithm with Nesterov's acceleration technique. Numerical simulations using real PAWR data show the effectiveness of the proposed method compared to conventional two-dimensional methods.

It is an important task to accurately track the target tumor with respiratory movement during radiation therapy. X-ray imaging technique is capable of observing the internal organ motion. However, superimposed tissues and structures in X-ray images decrease tumor localization accuracy. This paper presents a target extraction method based on hidden Markov model (HMM) to enhance the target tumor in X-ray images for improving the tumor tracking accuracy. We first simulate possible combinations of image intensities of target objects as hidden states and observable X-ray image intensities as output symbol in HMM by using digitally reconstructed radiographs generated from four-dimensional X-ray computed tomography. Subsequently, the transition dynamics of the hidden states and output symbols is estimated by applying Baum-Welch algorithm to a training dataset. The transition sequence of the hidden states is inversely estimated from the observed X-ray image sequence by using Viterbi algorithm, and then the transition sequence is finally decomposed into the subset image sequences. Experimental results demonstrated that tracking performance of the proposed method is superior to that of conventional tumor enhancement method and raw images. Therefore, the proposed method has potential for contributing to effectively observe internal organ motion.

In this paper, we show the effectiveness to use not only canny edge images but also hand contour images for hand pattern recognition by convolutional neural network (CNN). Hand contour images are binary images with only hand shapes and fingers information including fingers curled information. These hand contour images are generated using the colored glove which we proposed in the previous studies. In the experiments of hand pattern recognition, we investigate recognition accuracy by cross-validation method. Learning model using CNN consists of 4 convolution layers and 4 pooling layers. Moreover, Network In Network (NIN) is adopted as a convolution method. Test dataset is composed of only canny edge images, taking account of applicability for sign language recognition and so on. On the other hand, training dataset is composed of combination of canny edge images and hand contour images. Through the recognition experiments, we confirm the effectiveness of combining hand contour images in training dataset. The highest average recognition accuracy is 96.2% when combination rate of canny edge images and hand contour images is 50:50. This value is 6.9% higher, compared with the case combination rate is 100:0.

In this paper, a simple method based on Genetic Algorithm (GA) is proposed to evolve Block-Based Neural Network (BbNN) model. A BbNN consists of a 2-D array of memory-based modular component NNs with flexible structures and internal configuration that can be implemented in reconfigurable hardware such as a field programmable gate array (FPGA). The network structure and the weights are encoded in bit strings and globally optimized using the genetic operators. Asynchronous BbNN (ABbNN), which is a new model of BbNN, suggests high-performance BbNN by utilizing parallel computation and pipeline architecture. ABbNN's operating frequency is stable for all scales of the network, while conventional BbNN's is decreasing according to the network size. However, optimization by the genetic algorithm requires more iterations to find a solution with increasing problem space and the memory access in GA operation is one of the causes degrading the performance. ABbNN optimized with the proposed evolutionary algorithm is applied on general classifiers to verify the effectiveness with increasing problem space. The proposed method is confirmed by experimental investigations and compared with the conventional genetic algorithm.

This paper extended PMRL as the non-communicative and theoretical method for two agents, and proposed PLA as the method to be able to force agents to learn cooperative behavior for any number of agents. In addition, this paper adds the theoretic explanation for PLA that all agents achieve all purposes without spending the largest times. Concretely PLA forces each agent to avoid the more difficult purposes requiring many time to be reached by limiting the purpose which it can achieve, and it forces the agents to learn cooperative policy as achieving the appropriate purpose among the limited purposes. The experimental results in this paper derive that (1) PLA enables the agents to learn cooperative policy in the two grid world problems for three and five agents, and (2) PLA can force all agents to achieve all purposes in the problems with the minimum time.

Product price generally tends to decline as time proceeds. However, in Electronic commerce business (hereafter called “EC”), it sometimes increases influenced by supply and demand conditions; therefore, it is difficult to predict the future price. On the other hand, profit rate does not grossly fluctuate among companies running the same business. Focusing on this point, this study analyzes the tendencies of gross profit using the actual purchase and selling data of a certain EC operator, and also explores for the possibility of determining the optimal price by using these tendencies.

Recommender system is an information-filtering tool used in solving the problem that the user's preference in information overload. In recent years, with the development of deep learning, more and more research has begun to try the combination of deep learning and recommender system, and achieved good results. In this paper, we propose a novel probabilistic model. This model integrates a deep neural network model for extracting image shape features into the apparel goods recommender system to achieve good performance.

Conditional search system in employment supporting website is well used in Japan, if college or university students search place of employment. However, this system is not possible to respond flexibly to user’s undefined preference, and recommends only a few place of employment with strict conditions setup.

This paper proposes Development of the recommendation system of Place of Employment for Students with Decision rules in Rough Set using the condition relaxation based on user’s aptitude. First, this proposal system changes the attribute value of target place of employment into qualitative values. Next the system searches the place of employment which corresponds with the target user’s absolute conditions. When the number of search is not enough, the system eases conditions and searches additional places of employment, which are estimated by standardization Euclid distance, and which are selected by aptitude judgment. Finally, this system extracts decision rules of rough set based on the samples' judgement, and evaluates and shows places of employment, which are sorted in the descending order of the evaluation value. Through practical experiment, the proposal system recommended Place of Employment for Students more effectively than the usual Condition search.

In the development of the IoT systems, because it is necessary to test in various combinations, it takes much cost and time to execute tests for each combination and to analyze failures. In this paper, we have proposed a testing environment of the IoT systems to solve this issue. The proposed testing environment enables the IoT systems to work on the PC using the emulation technologies and testers can easily monitor inputs and outputs of each IoT device. The environment also automates combination tests and facilitates to manage the files for the automated test by managing functions to be tested and configurations to be tested separately. We have developed a prototype of the proposed testing environment for an air conditioning IoT system and verified that the environment had reduced the time to execute combination tests and to analyze failures.

Hypertension is a main factor of lifestyle diseases, and monitoring daily blood pressure (BP) could help prevent these diseases. To monitor BP daily, non-contact methods have been developed, but these methods are difficult to implement because they are used in specific areas. Here, an electric circuit model, as a novel model for BP estimation, was constructed by applying thermo-hue hemodynamic analysis (THHDA). Resultantly, the systolic blood pressure (SBP) estimated by the model could grasp characteristics of varied BP in the Valsalva maneuver, and the root mean squared error (RMSE) between estimated and measured SBP was as low as 17.571 mmHg.