Quantitative analysis of NaI(Tl) gamma-ray spectrometry using an artificial neural network

Kim, Jinhwan,Lim, Kyung Taek,Kim, Junhyeok,Kim, Chang-jong,Jeon, Byoungil,Park, Kyeongjin,Kim, Giyoon,Kim, Hojik,Cho, Gyuseong Elsevier BV * North-Holland 2019 Nuclear Instruments & Methods in Physics Research. Vol. No.

<P><B>Abstract</B></P> <P>In this manuscript, we propose an algorithm based on an artificial neural network (ANN) for the analysis of the NaI(Tl) gamma-ray spectra with radioisotope (RI) mixtures to identify RIs and determine the relative activity levels of the identified RIs. The ANN was trained based on the spectra that were generated by synthesizing previously identified spectra from single RIs, considering the characteristics of the measurement environments, such as gain shift effects and statistical fluctuations in the spectrum. The proposed ANN was evaluated through several measured spectra that contained up to six certified reference materials for a quantitative analysis. We also evaluated the shift in the spectra due to temperature variations in the range of 0–50 °C and the low-count spectra with a one-second acquisition period. These results were compared with those from an ANN trained through simulated spectra to emphasize the importance of acquiring a high-quality training dataset. In addition, we show that complex low-resolution spectra can be accurately analyzed with the proposed ANN under various scenarios, in which the maximum root mean square error was found to be 2.8%.</P>

Zorro의 연관키 차분특성을 이용한 키 복구 공격 및 PGV-Zorro의 충돌쌍 공격

김기윤(Giyoon Kim),박은후(Eunhu Park),이종혁(Jonghyeok Lee),장성우(Sungwoo Jang),김지훈(Jihun Kim),김한기(Hangi Kim),김종성(Jongsung Kim) 한국정보보호학회 2018 정보보호학회논문지 Vol.28 No.5

블록암호 Zorro는 AES와 비슷한 연산과정을 거치나 S-box 개수를 줄여 부채널 공격에 대비한 마스킹을 구현비용을 줄일 수 있게 설계되었다. 하지만 마스터키를 라운드키로 그대로 사용하기 때문에 연관키 차분공격 관점에서 취약하다. 본 논문에서는 Zorro의 연관키 차분특성을 이용해 키 복구 공격을 보인다. 또한 안전성이 증명된 12가지의 PGV 모델을 Zorro를 기반으로 할 때 블록암호의 연관키 차분 특성이 해시함수 충돌쌍 공격에 어떻게 활용될 수 있는지 설명하고 실제 충돌쌍을 제시한다. The block cipher Zorro is designed to reduce the implementation cost for side-channel countermeasure. It has a structure similar to AES, but the number of S-Boxes used is small. However, since the master key is used as the round key, it can be vulnerable to related key attacks. In this paper, we show key recovery attacks on Zorro using related-key differential characteristics. In addition, the related key differential characteristics are fatal when Zorro is used as the base block cipher of the hash function. In this paper, we describe how these characteristics can be linked to collision attacks in the PGV models.

Development of Dual-Arm Anticancer Drug Compounding Robot and Preparation System with Adaptability and High-Speed

Nam, Giyoon,Kim, Young Joo,Kim, Yun Jung,Kim, Yeoun Jae,Seo, Jung Ae,Kim, Kyunghwan,Kim, Kwang Gi International Society for Simulation Surgery 2016 Journal of International Society for Simulation Su Vol.3 No.2

Aim Robots are able to increase safety for pharmacy staff as separating from toxicity of anti-cancer drugs. For patient safety, it would provide right dose of the drugs. Additionally, it can reduce price of the drugs. Therefore, in this study, a novel compounding anticancer drugs robot system (Dupalro) was developed. Methods We used the robot system, Motoman dual-arm robot from YASKAWA, Japan and medications which are adapted for the robot were constructed. In order to develop a process of compounding anticancer drugs, information about five medications that are required to make anticancer drugs in hospitals was used. Results System for the five types of medications was constructed, and relating procedures for anticancer drugs compounding robot were developed. Conclusion Dupalro successfully was able to not only provide incremental safety and efficiency for both patients and pharmacy staff, but also decrease price of anticancer drugs.

Minimization for Temperature Dependence of the SiPM Adapting Count Rate Compensation and Operation Sequence

Giyoon Kim,Sanggeol Jeong,Jaeyeong Jang,Heejun Chung,Myungsoo Kim 한국방사성폐기물학회 2022 한국방사성폐기물학회 학술논문요약집 Vol.20 No.2

In emergency situations such as nuclear accidents or terrorism, radioactive and nuclear materials can be released by some environmental reasons such as the atmosphere and underground water. To secure the safety of human beings and to respond appropriately emergency situation, it is required to designate high and low dose rate regions in the early stages by analyzing the location and radioactivity of sources through environmental radiation measurement. This research team has developed a small gamma probe which is featured by its geometrical accessibility and higher radiation sensitivity than other drone detectors. A plastic scintillator and Silicon Photomultiplier (SiPM) were applied to the probe to optimize the wireless measurement condition. SiPM has a higher gain (higher than 106) and lower operating voltage (less than 30 V) compared to a general photodiode. However, the electronic components in the SiPM are sensitively affected by temperature, which causes the performance degradation of the SiPM. As the SiPM temperature increases, the breakdown voltage (VBD) of the SiPM also increases, so the gain must be maintained by applying the appropriate VBD. Therefore, when the SiPM temperature increases while the VBD is fixed, the gain decreases. Thus, the signal does not exceed the threshold voltage (VTH) and the overall count is reduced. In general, the optimal gain is maintained by cooling the SiPM or through a temperature compensation circuit. However, in the developed system, the hardware correction method such as cooling or temperature compensation circuit cannot be applied. In this study, it was confirmed that the count decreased by up to 20% according to the increase in the temperature of the SiPM when the probe was operated at room temperature (26°C). We propose methods to calibrate the total count without cooling device or compensation circuit. After operating the probe at room temperature, the first measured count is set as the reference value, and the correction factor is derived using the tendency of the count to decrease as the temperature increases. In addition, since this probe is used for environmental radiation monitoring, periodic measurements are more suitable than continuous measurements. Therefore, the temperature of the probe can be maintained by adding a power saving interval to the operation sequence of the probe. These two methods use the operation sequence and measurement data, respectively. Thus, it is expected to be the most effective method for the current system where the temperature compensation through hardware is not possible.

페르소나와 시나리오를 활용한 모바일 연구방법론 제안

김지윤(Giyoon Kim),이준민(Joonmin Lee),홍나희(Nahi Hong),이준환(Joonhwan Lee) 한국HCI학회 2016 한국HCI학회 학술대회 Vol.2016 No.1

Effects of Fish-Meal Application on the Growth of Chinese Cabbages (Brassica Campestris L.) and Soil Chemical Properties

Jeomsoo Kim,Heonil Kang,Giyoon Kwon,Namsook Park,Donggeun Kim,Insoo Choi 한국토양비료학회 2019 한국토양비료학회지 Vol.52 No.3

This study was conducted to get the basic data for the eco-friendly cultivation of Chines cabbages that high nitrogen requirements by evaluate the effect of the pollack fish-meal incorporation on growth properties of cabbage and soil chemistry. The study was conducted as pot (1/2,000a) test at the Miryang in Korea from May to August 2017. The treatment was divided into two groups; the amount of fish-meal treated were based by comparing the standard application ratio of nitrogen for Chinese cabbage with the nitrogen content of the fish-meal, and the chemical properties and the growth of the cabbage were compared by the treatments of 50%, 100%, 150%, 200% and 250%. The results obtained are following; the ammonia gas was produced from the after 2nd day of incorporated, and the amount of ammonia gas generated was geometrically correlated with the amount of fish processed. The soil pH level after study was lowered to about the pH 5.1 from 32.1 g/pot (1/2,000a). Due to the fish-meal increased the EC, content of Av. P₂O5 acid and inorganic nitrogen, and the pH level did not decrease even when the inorganic nitrogen increased at NO₃<SUP>-</SUP>‒N content of 5.4 mg/kg and NH₄<SUP>+</SUP>‒N was higher than 3.3 mg/kg. The SPAD value at the time of cabbage harvest was positive correlation with the amount of fish-meal treated, and the biomass was similar to that of SPAD. The biomass of Chinese cabbage in fish-meal plots were higher than that of conventional treatments, and the weight of the headed cabbage was the highest in T4 (64.1 g/pot). The number of cabbage leaf was the highest in T3 (53.4 g/pot). Based on the above results, it is thought that if the C/N ratio adjusted plant organic resources, methods of mixing and technologies for minimizing pH drop are supplemented, it will be excellent eco-friendly organic resources instead of synthetic nitrogen.

Machine learning algorithm for localization of nuclear materials based on gamma probe data to verify the denuclearization

Jeong Sanggeol,Kim Dong Yeung,Kim Giyoon,Lee Jun,Chung Heejun,Kim Myungsoo 한국물리학회 2023 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.83 No.12

The denuclearization verifcation process involves the localization of nuclear materials in the area of nuclear inspection. Various methodologies based on detector measurement using CsI(Tl), NaI(Tl) scintillators and Geiger–Müller (GM) counters have been studied to localize a nuclear material, but they are not suitable for application to a wide outdoor range. The Korea Institute of Nuclear Nonproliferation and Control (KINAC) has developed a plastic scintillator-based small gamma-ray instrument (probe). In this study, artifcial intelligence-based machine learning was applied to localize radioactive material based on probe measurement values. A localization algorithm model based on a Deep Neural Network (DNN) and Multiple Linear Regression (MLR) which are most used among various machine learning and deep learning algorithms was created. Then, the radioactive material was localized based on the measured value and compared with MCNP6-based simulation data. The performance of the DNN and MLR algorithms was evaluated through a coefcient of determination (R2 ) and Root Mean Square Error (RMSE). The results for the R2 and RMSE of the DNN algorithms model are 0.9488 and 3.5734 m. The R2 and RMSE of the MLR algorithm model are 0.8496 and 7.2452 m.

Strain Sensor를 이용한 Fuel Sender 개발

김병규(Byungkyu Kim),김태송(Taesong Kim),강기윤(Giyoon Kang),이진희(Jinhee Lee),문성욱(Sung Moon),박종오(Jong-oh Park),홍예선(Yehsun Hong) Korean Society for Precision Engineering 2001 한국정밀공학회지 Vol.18 No.6

짧은 오류정정부호를 위한 OSD 알고리즘의 복잡도 분석

김창현(Chang-Hyeon Kim),감동윤(Dongyun Kam),김석기(Seokki Kim),박기윤(Giyoon Park),이영주(Youngjoo Lee) 대한전자공학회 2020 대한전자공학회 학술대회 Vol.2020 No.8

In this paper, we analyze the hardware complexity of the ordered statistics decoding (OSD) algorithm, which is considered as one of the future shortlength error-correction codes due to its MLachieving performance. For fair comparisons, the conventional OSD method is compared to the 5G polar code associated with the same length and rate. By considering the practical systems, in addition, we perform the quantization for each decoding method, presenting the design challenges for realizing OSD approaches in terms of the decoding complexity.

macOS 환경에서의 Huawei 및 Apple 스마트폰 암호화 백업 데이터 복호화 및 아티팩트 분석

강수진(Soojin Kang),김기윤(Giyoon Kim),김소람(Soram Kim),김종성(Jongsung Kim) 한국디지털포렌식학회 2021 디지털 포렌식 연구 Vol.15 No.4

사용자와 관련한 다양한 데이터를 저장하는 스마트폰은 디지털 포렌식 관점에서 주요 분석 대상이다. 스마트폰의 보안 문제로 수사를 위한 스마트폰 데이터의 획득이 어려운 경우, 제조사 백업 서비스를 통한 데이터 획득 방식을 활용할 수 있다. 백업 기능은 제조사에서 사용자의 편의를 위해 데이터 보관 및 이동 등이 용이하도록 제공하는 서비스로 스마트폰 내의 주요 데이터를 PC나 다른 스마트폰으로 옮길 수 있다. 제조사에 따라 관리하는 데이터의 종류 및 형태가 다르며, 암호화 기능을 제공하여 한층 강화된 데이터 보안을 제공하기도 한다. 따라서 백업 데이터 활용 시 제조사별 백업 데이터 구조 및 암호화 기능 사용 여부에 따른 획득 가능한 데이터 종류를 파악해야 한다. 만일 암호화 기능이 적용된 경우, 데이터 복호화 방안 연구가 선행되어야 한다. 또한, 기존 연구 결과가 존재하더라도 백업 프로그램 업데이트 시 백업 데이터 구조 및 암호화 과정이 변경될 수 있어 지속적인 연구가 필요하다. 본 논문에서는 최신 macOS 환경에서 스마트폰 제조사인 Huawei와 Apple의 백업 데이터를 분석하였다. 각 제조사가 제공하는 백업 프로그램을 사용하여 백업된 데이터의 분석을 통해 상세한 암호화 과정을 밝혔다. 이를 기반으로 백업된 데이터를 모두 복호화하였으며, 암호화에 사용된 사용자 입력 패스워드의 복구 가능성을 제시하였다. 이후 복호화된 데이터 중 디지털 포렌식 관점에서 유용한 아티팩트를 선별하고 정리하였다. Smartphones that store various data related to users are a major analysis target in a digital forensic investigation. When it is difficult to obtain smartphone data for investigation due to security problems, a data acquisition method through a manufacturer"s backup service can be used. The backup function is a service provided by the manufacturer to facilitate data storage and movement for the convenience of users, and it is possible to move the main data in the smartphone to a PC or another smartphone. The types and forms of managed data differ depending on the manufacturer, and it also provides enhanced data security by providing an encryption function. Therefore, when using backup data, it is necessary to identify the types of data that can be obtained depending on the backup data structure by manufacturer and whether or not the encryption function is used. If an encryption function is applied, a study on data decryption method should be preceded. In addition, even if preceding studies results exist, the backup data structure and encryption process may be changed when updating the backup program, so subsequent studies are required. In this paper, we analyzed the backup data of smartphone manufacturers Huawei and Apple in the latest macOS environment. We revealed the detailed encryption process through the analysis of the backed up data using the backup program provided by Huawei and Apple manufacturers. Based on this, we decrypted all the backed up data, and suggested the possibility of recovering the user input password used for encryption. Then, we selected and organized useful artifacts from a digital forensics perspective among decrypted backed up data.