• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.409-416
• /
• 2022

This paper relates to the blind noise separation method of time-delayed convolutive mixed signals. Since the mixed model of acoustic signals in a closed space is multi-channel, a convolutive blind signal separation method is applied and time-delayed data samples of the two microphone input signals is used. For signal separation, the mixing coefficient is calculated using an inverse model rather than directly calculating the separation coefficient, and the coefficient update is performed by repeated calculations based on secondary statistical properties to estimate the speech signal. Many simulations were performed to verify the performance of the proposed blind signal separation. As a result of the simulation, noise separation using this method operates safely regardless of convolutive mixing, and PESQ is improved by 0.3 points compared to the general adaptive FIR filter structure.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.1
• /
• pp.113-119
• /
• 2023

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.6
• /
• pp.357-365
• /
• 2022

Refrigerant mischarging is one of the most frequently occurring failure modes in air conditioners, and both undercharging and overcharging degrade cooling performance. Therefore, it is important to accurately determine the amount of charged refrigerant. In this study, a support vector machine (SVM) model was developed to multi-classify the refrigerant mischarge through steady-state identification via fuzzy clustering techniques. For steady-state identification, a fuzzy clustering algorithm was applied to the air conditioner operation data using the difference between moving averages. The identification results using the proposed method were compared with those using existing steady-state determination techniques studied through the inversed Fisher's discriminant ratio (IFDR). Subsequently, the main features were selected using minimum redundancy maximum relevance (mRMR) considering the correlation among candidate features, and an SVM multi-classification model was devised using the derived features. The proposed method achieves satisfactory accuracy and robustness from test data collected in the new domain.

• Journal of the Korea Society for Simulation
• /
• v.18 no.4
• /
• pp.207-217
• /
• 2009

We propose a two-stage Distributed Denial of Service (DDoS) defense system, which can protect a given subnet by serving existing flows and new flows with a different priority based on IP history information. Denial of Service (DoS) usually occurs when the resource of a network node or link is limited and the demand of the users for that resource exceeds the capacity. The objective of the proposed defense system is to provide continued service to existing flows even in the presence of DDoS attacks. The proposed scheme can protect existing connections effectively with a smaller memory size by reducing the monitored IP address set through sampling and per-prefix history management. We evaluate the performance of the proposed scheme through simulation.

• Journal of Internet of Things and Convergence
• /
• v.9 no.2
• /
• pp.71-76
• /
• 2023

In this paper, we propose a smoothed received signal strength indicator (RSSI)-based distance estimation using deep neural network (DNN) for accurate distance estimation in an environment where a single receiver is used. The proposed scheme performs a data preprocessing consisting of data splitting, missing value imputation, and smoothing steps to improve distance estimation accuracy, thereby deriving the smoothed RSSI values. The derived smoothed RSSI values are used as input data of the Multi-Input Single-Output (MISO) DNN model, and are finally returned as an estimated distance in the output layer through input layer and hidden layer. To verify the superiority of the proposed scheme, we compared the performance of the proposed scheme with that of the linear regression-based distance estimation scheme. As a result, the proposed scheme showed 29.09% higher distance estimation accuracy than the linear regression-based distance estimation scheme.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.242-253
• /
• 2022

When the source waveform is unknown, the Green's function can be estimated by Ray-based Blind Deconvolution (RBD) based on the simple array signal processing. In previous papers, RBD was successfully demonstrated using simulation and experiments in shallow water environment. In this paper, we investigate the applicability of RBD for a long-range communication (e.g., 30 km, 60 km, and 90 km) in a deep water environment (1,000 m ~), using experimental data conducted in the east of Pohang, South Korea, in October 2018. Data results are presented to demonstrate Green's function estimation of a communication signal (2.2 kHz ~ 2.9 kHz) using a 16-element, 42-m long vertical array. The results show that the Green's function estimated from RBD is comparable to that of matched filter result. Additional communication performance at a maximum range of 90 km will be also presented.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.3
• /
• pp.339-349
• /
• 2017

This paper proposes a four-stage algorithm for detecting lanes on a driving car. In the first stage, it extracts region of interests in an image. In the second stage, it employs a median filter to remove noise. In the third stage, a binary algorithm is used to classify two classes of backgrond and foreground of an input image. Finally, an image erosion algorithm is utilized to obtain clear lanes by removing noises and edges remained after the binary process. However, the proposed lane detection algorithm requires high computational time. To address this issue, this paper presents a parallel implementation of a real-time line detection algorithm on a multi-core architecture. In addition, we implement and simulate 8 different processing element (PE) architectures to select an optimal PE architecture for the target application. Experimental results indicate that 40×40 PE architecture show the best performance, energy efficiency and area efficiency.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.45.2-46
• /
• 2021

The Hayabusa 2 mission target asteroid (162173) Ryugu is a near-Earth, carbonaceous (C-type) asteroid. Before the arrival, this asteroid is expected to be covered with mm- to cm- sized grains through the thermal infrared observations [1]. These grains are widely understood to be formed by past impacts with other celestial bodies and fractures induced by thermal fatigue [2]. However, the close-up images by the MASCOT lander showed lumpy boulders but no abundant fine grains [3]. Morota et al. suggested that there would be submillimeter particles on the top of these boulders but not resolved by Hayabusa 2's onboard instruments [4]. Hence, we conducted polarimetry of Ryugu to investigate microscopic grain sizes on its surface. Polarimetry is a powerful tool to estimate physical properties such as albedo and grain size. Especially, it is known that the maximum polarization degree (Pmax) and the geometric albedo (pV) show an empirical relationship depending on surface grain sizes [5]. We observed Ryugu from UT 2020 November 30 to December 10 at large phase angles (ranging from 78.5 to 89.7 degrees) to derive Pmax. We modified TRIPOL (Triple Range Imager and POLarimeter, [6]) to attach to the 1.8-m telescope at the Bohyunsan Optical Astronomy Observatory (BOAO). With this instrument, we observed the asteroid and determined linear polarization degrees at the Rc-band filter. We obtained sufficient data sets from 7 nights at this observatory to determine the Pmax value, and collaborated with other observatories in Japan (i.e., Hokkaido University, Higashi-Hiroshima, and Nishi-Harima) to acquire linear polarization degrees of the asteroid from total 24 nights observations with large phase angle coverage (From 28 to 104 degrees). The observational results have been published in Kuroda et al. (2021) [7]. We thus found the dominance of submillimeter particles on the surface of Ryugu from the comparison with other meteorite samples from the campaign observation. In this presentation, we report our activity to modify the TRIPOL for the 1.8-m telescope and the polarimetric performance. We also examine the rotational variability of the polarization degree using the TRIPOL data.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.115-117
• /
• 2021

Currently, as IoT technology develops, monitoring systems are being used in various fields, and image processing is being used in various forms. Image data causes noise due to various causes during the transmission and reception process, and if it is not removed, loss of image information or error propagation occurs. Therefore, denoising images is essential. Typical methods of eliminating Salt and Pepper noise in images include AF, MF, and A-TMF. However, existing methods have the disadvantage of being somewhat inadequate in high-density noise. Therefore, in this paper, we propose an algorithm for determining noise for Salt and Pepper denoising and replacing the central pixel with an original pixel if it is non-noise, and processing the filtering mask by segmenting and averaging it in eight directions. We evaluate the performance by comparing and analyzing the proposed algorithms with existing methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1773-1793
• /
• 2023

Automated machine learning, often referred to as "AutoML," is the process of automating the time-consuming and iterative procedures that are associated with the building of machine learning models. There have been significant contributions in this area across a number of different stages of accomplishing a data-mining task, including model selection, hyper-parameter optimization, and preprocessing method selection. Among them, preprocessing method selection is a relatively new and fast growing research area. The current work is focused on the recommendation of preprocessing methods, i.e., feature subset selection (FSS) algorithms. One limitation in the existing studies regarding FSS algorithm recommendation is the use of a single learner for meta-modeling, which restricts its capabilities in the metamodeling. Moreover, the meta-modeling in the existing studies is typically based on a single group of data characterization measures (DCMs). Nonetheless, there are a number of complementary DCM groups, and their combination will allow them to leverage their diversity, resulting in improved meta-modeling. This study aims to address these limitations by proposing an architecture for preprocess method selection that uses ensemble learning for meta-modeling, namely AutoFE-Sel. To evaluate the proposed method, we performed an extensive experimental evaluation involving 8 FSS algorithms, 3 groups of DCMs, and 125 datasets. Results show that the proposed method achieves better performance compared to three baseline methods. The proposed architecture can also be easily extended to other preprocessing method selections, e.g., noise-filter selection and imbalance handling method selection.