• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.688-693
• /
• 2022

In this paper, channel propagation path loss was measured in building corridors for frequency bands of 3.7 GHz and 28 GHz, which are used in 5G mobile communication, and compared and analyzed with CI (Close-In) and FI (Floating-Intercept) channel models. To measure the propagation path loss, the measurement was performed while moving the receiver (Rx) from the transmitter (Tx) by 10 m. As a result of the measurement, the PLE (Path Loss Exponent) values of the CI model at 3.7 GHz and 28 GHz were 1.5293 and 1.7795, respectively, and the standard deviations were analyzed as 9.1606 and 8.5803, respectively. In the FI model, 𝛼 values were 79.5269 and 70.2012, 𝛽 values were -0.6082 and 1.2517, respectively, and the standard deviations were 5.8113 and 4.4810, respectively. In the analysis results through the CI model and the FI model, the standard deviation of the FI model is smaller than that of the CI model, so it can be seen that the FI model is similar to the actual measurement result.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.456-457
• /
• 2021

Recently, in order to more accurately provide indoor location-based services, technologies using Wi-Fi fingerprints and deep learning are being studied. Among the deep learning models, an RNN model that can store information from the past can store continuous movements in indoor positioning, thereby reducing positioning errors. When using an RNN model for indoor positioning, the collected training data must be continuous sequential data. However, the Wi-Fi fingerprint data collected to determine specific location information cannot be used as training data for an RNN model because only RSSI for a specific location is recorded. This paper proposes a region clustering technique for sequential input data generation of RNN models based on Wi-Fi fingerprint data.

• The KIPS Transactions:PartB
• /
• v.19B no.3
• /
• pp.177-182
• /
• 2012

We propose effective motion and observation models for the position of a WiFi-equipped smartphone user in large indoor environments. Three component motion models provide better proposal distribution of the pedestrian's motion. Our Gaussian interpolation-based observation model can generate likelihoods at locations for which no calibration data is available. These models being incorporated into the particle filter framework, our WiFi fingerprint-based localization algorithm can track the position of a smartphone user accurately in large indoor environments. Experiments carried with an Android smartphone in a multi-story building illustrate the performance of our WiFi localization algorithm.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.458-459
• /
• 2021

Recently, researches using deep learning technology based on Wi-Fi fingerprints have been conducted for accurate services in indoor location-based services. Among the deep learning models, an RNN model that can store information from the past can store continuous movements in indoor positioning, thereby reducing positioning errors. At this time, continuous sequential data is required as training data. However, since Wi-Fi fingerprint data is generally managed only with signals for a specific location, it is inappropriate to use it as training data for an RNN model. This paper proposes a path generation method through prediction of a moving path based on Wi-Fi fingerprint data extended to region data through clustering to generate sequential input data of the RNN model.

• Journal of Food Hygiene and Safety
• /
• v.29 no.4
• /
• pp.299-304
• /
• 2014

The dose-response models are important for the quantitative microbiological risk assessment (QMRA) because they would enable prediction of infection risk to humans from foodborne pathogens. In this study, we performed a comprehensive literature review and meta-analysis to better quantify this association. The meta-analysis applied a final selection of 193 published papers for total 43 species foodborne disease pathogens (bacteria 26, virus 9, and parasite 8 species) which were identified and classified based on the dose-response models related to QMRA studies from PubMed, ScienceDirect database and internet websites during 1980-2012. The main search keywords used the combination "food", "foodborne disease pathogen", "dose-response model", and "quantitative microbiological risk assessment". The appropriate dose-response models for Campylobacter jejuni, pathogenic E. coli O157:H7 (EHEC / EPEC / ETEC), Listeria monocytogenes, Salmonella spp., Shigella spp., Staphylococcus aureus, Vibrio parahaemolyticus, Vibrio cholera, Rota virus, and Cryptosporidium pavum were beta-poisson (${\alpha}=0.15$, ${\beta}=7.59$, fi = 0.72), beta-poisson (${\alpha}=0.49$, ${\beta}=1.81{\times}10^5$, fi = 0.67) / beta-poisson (${\alpha}=0.22$, ${\beta}=8.70{\times}10^3$, fi = 0.40) / beta-poisson (${\alpha}=0.18$, ${\beta}=8.60{\times}10^7$, fi = 0.60), exponential (r=$1.18{\times}10^{-10}$, fi = 0.14), beta-poisson (${\alpha}=0.11$, ${\beta}=6,097$, fi = 0.09), beta-poisson (${\alpha}=0.21$, ${\beta}=1,120$, fi = 0.15), exponential ($r=7.64{\times}10^{-8}$, fi = 1.00), betapoisson (${\alpha}=0.17$, ${\beta}=1.18{\times}10^5$, fi = 1.00), beta-poisson (${\alpha}=0.25$, ${\beta}=16.2$, fi = 0.57), exponential ($r=1.73{\times}10{-2}$, fi = 1.00), and exponential ($r=1.73{\times}10^{-2}$, fi = 0.17), respectively. Therefore, these results provide the preliminary data necessary for the development of foodborne pathogens QMRA.

• Journal of the Korean Geotechnical Society
• /
• v.20 no.9
• /
• pp.47-56
• /
• 2004

A new fragmentation model, called the GRS (the generation model of weathered residual soils) model, was proposed in this study, This model could identify the formation of a residual soil. This model is based on the phenomena that as the soil was weathered more highly, soil particles were smaller and pores were more expanded simultaneously. The possibility of fragmentation, $P_F,$ which was based on the fractal theory, was introduced in this model. There were some fundamental notions in the GRS model that soil particles were generated as the rock is fragmented, and the fragmentation of the rock was performed step by step. The $P_F,$ of the rock was not constant at each fragmentation steps. As a result of application on the GRS model, there were more residue where $P_{Fi}s$ were small at any particle size. There was a S-shape of PSD curve at the concave shape of $P_{Fi},$ and the PSD curve goes to a gaped graded curve at the convex shape of $P_{Fi}.$ The shape of PSD curve was concave in the case of small $P_{Fi}s.$ The value of $P_{Fi}$ increased with the coefficient of uniformity $(C_u)$ and the fragmentation fractal dimension $(D_r),$ but had no relation with the coefficient of gradation $(C_C)$.

• 한국정보통신설비학회:학술대회논문집
• /
• 2008.08a
• /
• pp.360-363
• /
• 2008

현재 통신 시장의 상황은 유무선 통신서비스 시장의 포화, 이동통신 시장의 기존 유선전화 규모 초과, 그리고 유선통신의 서비스 사업자의 영역 확대 도모 등을 특징으로 한다. 유무선 통합 (Fixed Mobile Convergence, FMC) 서비스는 유선통신 사업자를 중심으로 한 비즈니스 모델로 사용자에게 유무선 통신망 종류에 상관없이 일관되고 끊김없는 서비스를 제공하는 것을 목표로 한다. 이동통신망까지 확대하여 고객 기반을 유지하고자 하는 유선사업자들은 FMC 서비스 중 하나의 방안으로 IMS (IP Multimedia Subsystems) 기반의 VCC(Voice Call Continuity) 기능에 주목하고 있다. VCC AS(Application Server)는 이종망 (WiFi-CDMA)간 Seamless 핸드오버기능을 수행하므로, WiFi 와 CDMA를 지원할 수 있는 듀얼단말을 이용하여 사용자가 WiFi 서비스 지역과 CDMA 서비스 지역간 이동시에도 Seamless 한 음성서비스를 제공한다. 이에 본 논문은 IMS/VCC 기반으로 음성 seamless 핸드오버 적용 사례(시범서비스)를 중심으로 유무선 통신사업자 상호 Win-Win을 추구할 수 있는 LG데이콤 특화 VCC 모델을 제시한다. 그리고 LG데이콤이 추구하는 차세대 서비스인 사용자 context 기반의 개인화된 서비스 제공을 위한 IMS 기반 통합 프로파일/인증/과금 연구 동향에 대해 간략히 소개한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.11
• /
• pp.473-478
• /
• 2022

Evaluation learning based on code testing is becoming a popular solution in programming education via Online judge(OJ). In the recent past, many papers have been published on how to detect plagiarism through source code similarity analysis to support OJ. However, deep learning-based research to support automated tutoring is insufficient. In this paper, we propose Input & Output side FiLM models to predict whether the input code will pass or fail. By applying Feature-wise Linear Modulation(FiLM) technique to GRU, our model can learn combined information of Java byte codes and problem information that it tries to solve. On experimental design, a balanced sampling technique was applied to evenly distribute the data due to the occurrence of asymmetry in data collected by OJ. Among the proposed models, the Input Side FiLM model showed the highest performance of 73.63%. Based on result, it has been shown that students can check whether their codes will pass or fail before receiving the OJ evaluation which could provide basic feedback for improvements.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.3
• /
• pp.113-120
• /
• 2017

In recent years, indoor localization has been researched for the improvement of its localization accuracy capability in Wi-Fi environment. The fingerprint and RF propagation models has been the main approach in determining indoor positioning. With the use of fingerprint, a low-cost, versatile localization system can be achieved without the use of external hardware. However, only a few research have been made on virtual access points (VAPs) among indoor localization models. In this paper, the idea of indoor localization system using fingerprint with the addition of VAP in Wi-Fi environment is discussed. The idea is to virtually add APs in the existing indoor Wi-Fi system, this would mean additional virtually APs in the network. The experiments of the proposed algorithm shows the positive results when 2VAPs are used compared with only APs. A combination of 3APs and 2VAPs in the 3rd case had the lowest average error of 3.99 among its 4 scenarios.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.341-348
• /
• 2022

Wi-Fi Sensing, which uses Wi-Fi technology to sense the surrounding environments, has strong potentials in a variety of sensing applications. Recently several advanced deep learning-based solutions using CSI (Channel State Information) data have achieved high performance, but it is still difficult to use in practice without explicit data collection, which requires expensive adaptation efforts for model retraining. In this study, we propose a Channel State Information Automatic Labeling System (CALS) that automatically collects and labels training CSI data for deep learning-based Wi-Fi sensing systems. The proposed system allows the CSI data collection process to efficiently collect labeled CSI for labeling for supervised learning using computer vision technologies such as object detection algorithms. We built a prototype of CALS to demonstrate its efficiency and collected data to train deep learning models for detecting the presence of a person in an indoor environment, showing to achieve an accuracy of over 90% with the auto-labeled data sets generated by CALS.