• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.309-316
• /
• 2018

Objectives: Both the valence and arousal components of affect are important considerations when managing mental healthcare because they are associated with affective and physiological responses. Research on arousal and valence analysis, which uses images, texts, and physiological signals that employ deep learning, is actively underway; research investigating how to improve the recognition rate is needed. The goal of this research was to design a deep learning framework and model to classify arousal and valence, indicating positive and negative degrees of emotion as high or low. Methods: The proposed arousal and valence classification model to analyze the affective state was tested using data from 40 channels provided by a dataset for emotion analysis using electrocardiography (EEG), physiological, and video signals (the DEAP dataset). Experiments were based on 10 selected featured central and peripheral nervous system data points, using long short-term memory (LSTM) as a deep learning method. Results: The arousal and valence were classified and visualized on a two-dimensional coordinate plane. Profiles were designed depending on the number of hidden layers, nodes, and hyperparameters according to the error rate. The experimental results show an arousal and valence classification model accuracy of 74.65 and 78%, respectively. The proposed model performed better than previous other models. Conclusions: The proposed model appears to be effective in analyzing arousal and valence; specifically, it is expected that affective analysis using physiological signals based on LSTM will be possible without manual feature extraction. In a future study, the classification model will be adopted in mental healthcare management systems.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.106-113
• /
• 2019

In this paper, we propose an algorithm to classify the received active pulse when the active sonar system is operated as a non-cooperative mode. The proposed algorithm uses CNN (Convolutional Neural Networks) which shows good performance in various fields. As an input of CNN, time frequency analysis data which performs STFT (Short Time Fourier Transform) of the received signal is used. The CNN used in this paper consists of two convolution and pulling layers. We designed a database based neural network and a pulse feature based neural network according to the output layer design. To verify the performance of the algorithm, the data of 3110 CW (Continuous Wave) pulses and LFM (Linear Frequency Modulated) pulses received from the actual ocean were processed to construct training data and test data. As a result of simulation, the database based neural network showed 99.9 % accuracy and the feature based neural network showed about 96 % accuracy when allowing 2 pixel error.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.1
• /
• pp.15-20
• /
• 2019

Inspection and measurement of surface quality is one of the most critical processes for manufacturing products such as semiconductor wafers, sapphire substrates, and display panels. The vibrations of the inspection and measurement devices are supposed to be the most dominant factors for severe measurement errors and longer measuring time. In this study, dynamic characteristics of an inspection and measurement device are analyzed through frequency response experiment and computer simulation to obtain parameters such as frequencies, magnitudes, mode shapes, and periods of vibrations. And then an active damper which consists of sensor, interface board, and actuator is designed based on the parameters to formulate the most effective reaction signal to suppress the vibrations which is generated by an interface board, and provided by an actuator. If the vibrations are measured by the sensor, the active damper immediately generates and provides the corresponding reaction signal to inspection and measurement device. The result shows that the active damper can suppress structural vibrations effectively and reduce measuring time of the device and enhance the productivity.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.46 no.11
• /
• pp.969-973
• /
• 2018

Avionics OFP runs on the mission computer and can be operated by interacting with several avionics equipments. In order to verify OFP SW, SIL having real avionics equipments or models is absolutely necessary. Therefore in many cases SIL is implemented concurrently with OFP developing, and only one SIL is provided to developers. So developers sometimes need an alternative to SIL for verifying requirements in the middle of development process. In this paper, we propose a single PC based STE SW that simulates interworking equipments and verifies OFP in a single PC environment without actual interworking equipments or SIL HW interfaces.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.3
• /
• pp.533-540
• /
• 2021

To evaluate the performance of a system, one-dimensional 3-neighborhood cellular automata(CA) based pseudo-random generators are widely used in many fields. Although two-dimensional CA and one-dimensional 5-neighborhood CA have been applied for more effective key sequence generation, designing symmetric one-dimensional 5-neighborhood CA corresponding to a given primitive polynomial is a very challenging problem. To solve this problem, studies on one-dimensional 5-neighborhood CA synthesis, such as synthesis method using recurrence relation of characteristic polynomials and synthesis method using Krylov matrix, were conducted. However, there was still a problem with solving nonlinear equations. To solve this problem, a symmetric one-dimensional 5-neighborhood CA synthesis method using a transition matrix of 90/150 CA and a block matrix has recently been proposed. In this paper, we detail the theoretical process of the proposed algorithm and use it to obtain symmetric one-dimensional 5-neighborhood CA corresponding to high-order primitive polynomials.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.123-130
• /
• 2021

With the recent developments in computer technology, there has been an increasing interest in the field of machine learning. This also has led to a significant increase in real business cases of machine learning theory in various sectors. In finance, it has been a major challenge to predict the future value of financial products. Since the 1980s, the finance industry has relied on technical and fundamental analysis for this prediction. For future value prediction models using machine learning, model design is of paramount importance to respond to market variables. Therefore, this paper quantitatively predicts the stock price movements of individual stocks listed on the KOSPI market using machine learning techniques; specifically, the reinforcement learning model. The DQN and A2C algorithms proposed by Google Deep Mind in 2013 are used for the reinforcement learning and they are applied to the stock trading strategies. In addition, through experiments, an input value to increase the cumulative profit is selected and its superiority is verified by comparison with comparative algorithms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.7
• /
• pp.857-865
• /
• 2020

This paper presents a method for code generation of UML sequence diagrams based on message patterns. In the sequence diagrams, it is shown that messages are some types of forms typically. This paper classifies according to type as three patterns, and construct meta-information for code generation analysing structural infomation for each patterns. The meta-message of structural information (MetaMessage) is stored in the MetaMessage datastore and the meta-method information from the MetaMessage is stored in the MetaMethod datastore. And then, the structural information of MetaClass and MetaObject is constructed in each datastore too. For each pattern, this paper presents a method for code generation based on the meta information of message patterns and the syntax of target progamming language. Also, branching and looping that has been seldom handled integratedly in the previous works are handled as same as the basic patterns by classifying the branching pattern and the looping pattern for code generation integratedly.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.6
• /
• pp.483-488
• /
• 2020

In recent years, the demand for wireless devices incorporating several wireless communication systems into one has been increasing in order to provide services that meet the diverse needs of consumers. Wireless devices consisting of various wireless communication systems require many frequency fixed filters. A frequency tunable filter can replace a number of frequency fixed filters in the wireless devices. If a frequency tunable filter is used in wireless systems, the system can be configured more efficiently. In this paper, a 3-pole frequency tunable BPF(bandpass filter) operating in the frequency band of 800 ~ 2400MHz is designed. In order to widen the operating frequency band, a tuning screw is designed to have a step and a linear motor is used to facilitate the adjustment of the tuning screw. The implemented frequency tunable BPF operates in the designed frequency range and has the maximum insertion loss of 2.82dB in the channel band and the minimum attenuation of 18.7dB at ± 50MHz frequency offset from the center frequency of the band.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.2
• /
• pp.149-157
• /
• 2020

Today, due to the large amount of data duplication caused by the increase in user data, various deduplication studies have been conducted. However, research on personal storage is relatively poor. Personal storage, unlike high-performance computers, needs to perform deduplication while reducing CPU and memory resource usage. In this paper, we propose an adaptive algorithm that selectively applies fixed size chunking (FSC) and whole file chunking (WFH) according to the file type and size in order to maintain the deduplication rate and reduce the load in personal storage. We propose an algorithm for minimization. The experimental results show that the proposed file system has more than 1.3 times slower at first write operation but less than 3 times reducing in memory usage compare to LessFS and it is 2.5 times faster at rewrite operation.

• Journal of Positioning, Navigation, and Timing
• /
• v.10 no.3
• /
• pp.169-177
• /
• 2021

In this paper, we propose a novel global navigation satellite system (GNSS) spoofing detection technique through an array antenna-based direction of arrival (DoA) estimation of satellite and spoofer. Specifically, we consider a sophisticated GNSS spoofing attack scenario where the spoofer can accurately mimic the multiple pseudo-random number (PRN) signals since the spoofer has its own GNSS receiver and knows the location of the target receiver in advance. The target GNSS receiver precisely estimates the DoA of all PRN signals using compressed sensing-based orthogonal matching pursuit (OMP) even with a small number of samples, and it performs spoofing detection from the DoA estimation results of all PRN signals. In addition, considering the initial situation of a sophisticated spoofing attack scenario, we designed the algorithm to have high spoofing detection performance regardless of the relative spoofing signal power. Therefore, we do not consider the assumption in which the power of the spoofing signal is about 3 dB greater than that of the authentic signal. Then, we introduce design parameters to get high true detection probability and low false alarm probability in tandem by considering the condition for the presence of signal sources and the proximity of the DoA between authentic signals. Through computer simulations, we compare the DoA estimation performance between the conventional signal direction estimation method and the OMP algorithm in few samples. Finally, we show in the sophisticated spoofing attack scenario that the proposed spoofing detection technique using OMP-based estimated DoA of all PRN signals outperforms the conventional spoofing detection scheme in terms of true detection and false alarm probability.