• Proceedings of the IEEK Conference
• /
• 2004.08c
• /
• pp.540-543
• /
• 2004

Address Resolution Protocol (ARP) cache poisoning is a MAC layer attack that can only be carried out when an attacker is connected to the same local network as the target machines. ARP is not a new problem, but wireless network introduces a new attack point and more vulnerable to the attack. The attack on wireless network cannot be detected by current detection tool installed on wired network. In order to detect the ARP poisoning attack, there must be a ARP poisoning detection tool for wireless LAN environment. This paper proposes linux-based ARP poisoning detection system equipped with wireless LAN card and Host AP device driver

• Convergence Security Journal
• /
• v.22 no.3
• /
• pp.25-32
• /
• 2022

Recently, the image processing industry has been activated as deep learning-based technology is introduced in the image recognition and detection field. With the development of deep learning technology, learning model vulnerabilities for adversarial attacks continue to be reported. However, studies on countermeasures against poisoning attacks that inject malicious data during learning are insufficient. The conventional countermeasure against poisoning attacks has a limitation in that it is necessary to perform a separate detection and removal operation by examining the training data each time. Therefore, in this paper, we propose a technique for reducing the attack success rate by applying modifications to the training data and inference data without a separate detection and removal process for the poison data. The One-shot kill poison attack, a clean label poison attack proposed in previous studies, was used as an attack model. The attack performance was confirmed by dividing it into a general attacker and an intelligent attacker according to the attacker's attack strategy. According to the experimental results, when the proposed defense mechanism is applied, the attack success rate can be reduced by up to 65% compared to the conventional method.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.3
• /
• pp.549-559
• /
• 2023

As mobile edge computing (MEC) is gaining attention as a core technology of 5G networks, edge AI technology of 5G network environment based on mobile user data is recently being used in various fields. However, as in traditional AI security, there is a possibility of adversarial interference of standard 5G network functions within the core network responsible for edge AI core functions. In addition, research on data poisoning attacks that can occur in the MEC environment of standalone mode defined in 5G standards by 3GPP is currently insufficient compared to existing LTE networks. In this study, we explore the threat model for the MEC environment using NWDAF, a network function that is responsible for the core function of edge AI in 5G, and propose a feature selection method to improve the performance of detecting data poisoning attacks for Leaf NWDAF as some proof of concept. Through the proposed methodology, we achieved a maximum detection rate of 94.9% for Slowloris attack-based data poisoning attacks in NWDAF.

• Journal of Preventive Medicine and Public Health
• /
• v.5 no.1
• /
• pp.111-114
• /
• 1972

There was an outbreak of food poisoning on the 17 October, 1970 among the primary school children who came from a rural area, Yeongi-gun, Choongcheongnam-do to Seoul City on an educational trip. Of the 199 children participating in the trip, 149 cases of food poisoning developed a 74.9% attack rate. The acute onset of symptoms, of abdominal pain, diarrhea, vomiting and headache which occurred 1-5 hours after eating their lunch suggests that the outbreak was due to staphylococcal food poisoning. The common source of food was identified as the lunch packed in a chip-box which were eaten on October 17 during the trip. Most probable kind of food of the lunch as the cause was the favoured fish paste. The lunch were prepared at restaurant A in Seoul City. One of the personnel of the restaurant had a unhealed cut wound on the third finger tip of the left hand, from which it was considered that the food was contaminated with Staphylococcus during preparation. The chance of multiplication of Staphylococcus to produce enterotoxin in the food might be existed during flavouring the food with some degree of heat, and also during about 10 hours elapsed before serving the food after preparation.

• Journal of agricultural medicine and community health
• /
• v.6 no.1
• /
• pp.33-41
• /
• 1981

To investigate the status of medical accident attack rate and medicare utilization during last 27months from 1st Oct., 1977 to 31st Dec. 1979 in the area under Sam Wha Medicare Insurance Union, the study was carried out through analyzing the medicare records of patients who were enrolled. "The medical accident" in this study was used as the meaning of the state that the people who have been treated morbid condition with insurance money. For the study, 2 doctors and one nurse were mobilized and the results are as follows: 1) The total number of the Medicare Insurance Union members among the 37,044 total population of the study area, (Hea-mi, Unsan, Eumam Myun) were 57, 35 composed of 3,000 males (52.3%) and 27, 35 females (47.7%) in 1977, 3,383 composed of 2,006 males (59.3%) 1,377 females (40.7%) in 1978, 2,573 composed of 1,437 males and 1,336 females (44.2%) in 1979. 2) Total number of medical accident attack cases were 6,774 case (partially overlapped the number of the 1977 with that of the 1978) and average annual medical accident attack rate per 1,000 population was 700.9. 3) Five major disease group in the past three years were disease of the respiratory system (177.7), disease of the digestive system (165.8), disease of the skin and subcutaneous(64.9), symptoms, signs and ill-defined conditions (64.6), any injury and poisoning (51.9). Mental disorders (32.6) was the 6th order disease group. 4) The order of the medical accident attack rate of age group per 1,000 population per year was the year group of 0~4(877.8), 45~64(832.6), 25~44(810.3), 5~14(495.1) 15~24(494.7) 65 and over (460.7). 5) Medical accident attack rate of age group per 1,000 population in the 5 major disease groups were the year group of 45~64 (100.0) in the mental disorders 0~4(525.1) in the disease of the respiratory system 45~64 (328.5) in the disease of the digestive system 0~4 (202.8) in the disease of the skin and subcutanous tissue, 25~44 (98.3) in the accidents and poisoning. 6) Monthly medical accidents attack rate were 87.0 in the winter (Dsc., Jan., Feb.) and 86.2 in the summer (Jun., July, Aug.). So the trend of the medical accidents attack rate during the year was bimount figure. 7) Monthly medical accidents attack rate in the major disease group were highest in Jan., Apr., (31.1) in the disease of the respiratory system, in Jan., Feb., Mar. (24.9) in the disease of the digestive system, in Jan., Jun., Aug. (9.8) in the injury and poisoning. 8) Duration of the treatment of the 93.1% of the total cases were within 5 days. 9) 299 cases (4.5%) of the total number of cases, 6,587 cases were referred to secondary and tertiary medicare facilities. 10) The order of the major 10 kinds of diagnosis of the disease, 6,587 cases during 27 months, were URI and chillness (1,063 cases, 16.1%), gastritis(830 cases, 12. 6%) dermatitis(360 cases, 5.5%), bronchitis(291 cases, 4.4%), neurosis (284 cases, 4.3%), contusion (165 cases, 2.5%), tooth extraction (157 cases, 2.4%), tonsillitis (109 cases, 0.7%), laceration (107 cases, 1.6%), neuralgia (105 cases, 1.6%), arthritis (104 cases, 1.6%), otitis media and mastoiditis (103 cases, 1.6%), so total case were 3,678 cases (55.9%).

• Journal of The Korean Society of Clinical Toxicology
• /
• v.19 no.1
• /
• pp.51-54
• /
• 2021

Chlorfenapyr is a widely used insecticide, that is very lethal if ingested. It exhibits delayed toxicity in which there are few symptoms at first which suddenly worsen after a few days. A 66-year-old female patient ingested about 90 mL of chlorfenapyr liquid hydrating agent (Chlofenapyr 10%) and showed stable vital signs with no specific symptoms and findings other than a mild fever, vomiting, and nausea. From the 3rd day of ingestion, creatine kinase was high, and rhabdomyolysis was suspected. From the 4th day of ingestion, pancreatic enzymes began to gradually increase. A diffusion-weighted image showed a multifocal high signal intensity in the white matter and corpus callosum area. On the 8th day after ingestion, she suffered a high fever and a heart attack and died. Thus, if a patient is suspected of taking chlorfenapyr, he/she needs active treatment and monitoring even if he/she does not exhibit any symptoms.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.79-80
• /
• 2021

본 논문은 MITM 공격기법을 역이용한 네트워크 보안 기술 및 구현 방안을 제시한다. MITM(Man In The Middle) 공격은 통신 경로 중간에 개입하여 양 단간의 통신 내용을 가로채거나 행위 제어를 수행하는 전통적인 해킹 방법으로 그 공격 기법을 역이용하여 네트워크 공격을 방어하는 보안기술 및 시스템 구현에 대해 기술한다. Linux 시스템을 이용하여 ARP Poisoning을 통해 양단간 통신 트래픽에 개입하며, Netfilter 및 Suricata를 이용하여 Network IDS/IPS 및 Firewall을 구현하였고, Contents 필터링 및 Anti-Virus 구현이 가능하며, 여러 기능을 확장하여 UTM(Unified Threats Management) 시스템을 구현하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.621-623
• /
• 2023

본 논문에서는 기계 학습 모델의 취약점과 대응책에 초점을 맞추어 적대적인 기계 학습 공격 및 방어 분야를 탐구한다. 신중하게 만들어진 입력 데이터를 도입하여 기계 학습 모델을 속이거나 조작하는 것을 목표로 하는 적대적 공격에 대한 심층 분석을 제공한다. 이 논문은 회피 및 독성 공격을 포함한 다양한 유형의 적대적 공격을 조사하고 기계 학습 시스템의 안정성과 보안에 대한 잠재적 영향을 조사한다. 또한 적대적 공격에 대한 기계 학습 모델의 견고성을 향상시키기 위해 다양한 방어 메커니즘과 전략을 제안하고 평가한다. 본 논문은 광범위한 실험과 분석을 통해 적대적 기계 학습에 대한 이해에 기여하고 효과적인 방어 기술에 대한 통찰력을 제공하는 것을 목표로 한다.

• International Journal of Computer Science & Network Security
• /
• v.22 no.5
• /
• pp.193-203
• /
• 2022

After the everyday use of systems and applications of artificial intelligence in our world. Consequently, machine learning technologies have become characterized by exceptional capabilities and unique and distinguished performance in many areas. However, these applications and systems are vulnerable to adversaries who can be a reason to confer the wrong classification by introducing distorted samples. Precisely, it has been perceived that adversarial examples designed throughout the training and test phases can include industrious Ruin the performance of the machine learning. This paper provides a comprehensive review of the recent research on adversarial machine learning. It's also worth noting that the paper only examines recent techniques that were released between 2018 and 2021. The diverse systems models have been investigated and discussed regarding the type of attacks, and some possible security suggestions for these attacks to highlight the risks of adversarial machine learning.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.666-669
• /
• 2024

연합학습은 원본 데이터를 공유하지 않고 모델을 학습할 수 있는 각광받는 프라이버시를 위한 학습방법론이다. 이를 위해 참여자의 데이터를 수집하는 대신, 데이터를 인공지능 모델 학습의 요소들(가중치, 기울기 등)로 변환한 뒤, 이를 공유한다. 이러한 강점에 더해 기존 연합학습을 개선하는 방법론들이 추가적으로 연구되고 있다. 기존 연합학습은 모델 가중치를 평균내는 것으로 참여자 간에 동일한 모델 구조를 강요하기 때문에, 참여자 별로 자신의 환경에 알맞은 모델 구조를 사용하기 어렵다. 이를 해결하기 위해 지식 증류 기반의 연합학습 방법(Knowledge Distillation-based Federated Learning)으로 서로 다른 모델 구조를 가질 수 있도록(Model Heterogenousity) 하는 방법이 제시되고 있다. 연합학습은 여러 참여자가 연합하기 때문에 일부 악의적인 참여자로 인한 모델 포이즈닝 공격에 취약하다. 수많은 연구들이 기존 가중치를 기반으로한 연합학습에서의 위협을 연구하였지만, 지식 증류 기반의 연합학습에서는 이러한 위협에 대한 조사가 부족하다. 본 연구에서는 최초로 지식 증류 기반의 연합학습에서의 모델 성능 하락 공격에 대한 위협을 실체화하고자 한다. 이를 위해 우리는 GMA(Gaussian-based Model Poisoning Attack)과 SMA(Sign-Flip based Model Poisoning Attack)을 제안한다. 결과적으로 우리가 제안한 공격 방법은 실험에서 최신 학습 기법에 대해 평균적으로 모델 정확도를 83.43%에서 무작위 추론에 가깝게 떨어뜨리는 것으로 공격 성능을 입증하였다. 우리는 지식 증류 기반의 연합학습의 강건성을 평가하기 위해, 새로운 공격 방법을 제안하였고, 이를통해 현재 지식 증류 기반의 연합학습이 악의적인 공격자에 의한 모델 성능 하락 공격에 취약한 것을 보였다. 우리는 방대한 실험을 통해 제안하는 방법의 성능을 입증하고, 결과적으로 강건성을 높이기 위한 많은 방어 연구가 필요함을 시사한다.