• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.524-530
• /
• 2023

This paper proposes an efficient algorithm to detect CAN (Controller Area Network) bus attack based on a lightweight CNN (Convolutional Neural Network), and an IDS(Intrusion Detection System) was designed, implemented, and verified with FPGA. Compared to conventional CNN-based IDS, the proposed IDS detects CAN bus attack on a frame-by-frame basis, enabling accurate and rapid response. Furthermore, the proposed IDS can significantly reduce hardware since it exploits only one convolutional layer, compared to conventional CNN-based IDS. Simulation and implementation results show that the proposed IDS effectively detects various attacks on the CAN bus.

• Journal of Industrial Convergence
• /
• v.21 no.10
• /
• pp.91-101
• /
• 2023

This study was attempted to identify the essence by confirming the experiences of students who participated in Habuta classes in their major subjects. From November 22, 2022 to December 6, 2022, eight nursing students were individually interviewed to collect data, and the data were analyzed by applying Colaizzi's phenomenological analysis methodology. The study's finding derived 6 categories. The specific categories were 'Coexistence of anxiety and expectation', 'Process of understanding and acquisition', 'Constant burden and discomfort', 'Progress and regression under pressure', 'Sense of security that is being recognized', 'Shift of thought to positivity'. In order for nursing students to participate in Havruta classes positively in the future, appropriate class design, active communication channels between professors and students, and strengthening of instructors' competence are required.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.137-145
• /
• 2023

In this paper, we propose system access control software that improves the survivability of the naval combat system while maintaining security and stability. The software was improved by improving the operating environment configuration and user authentication process of the system access control software that constitutes the naval combat system, defining the operating environment classification of the naval combat system, and applying a software parallel execution process considering the load rate. Through this, the waiting time required to configure the environment is shortened, providing rapid operation to the operator, and improving the fact that the naval combat system cannot be operated unless the environment configuration is completed even in emergency situations. In order to test the performance, a test environment was created by simulating the existing naval combat system, and the execution time for each operation category was measured and compared. Compared to the existing naval combat system, the execution time of the basic combat system was reduced by about 69.3%, the execution time of the combat system was reduced by about 54.9%, and the execution time of the integrated combat system was confirmed to be reduced by about 8.4%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3383-3397
• /
• 2023

Scene graphs serve as semantic abstractions of images and play a crucial role in enhancing visual comprehension and reasoning. However, the performance of Scene Graph Generation is often compromised when working with biased data in real-world situations. While many existing systems focus on a single stage of learning for both feature extraction and classification, some employ Class-Balancing strategies, such as Re-weighting, Data Resampling, and Transfer Learning from head to tail. In this paper, we propose a novel approach that decouples the feature extraction and classification phases of the scene graph generation process. For feature extraction, we leverage a transformer-based architecture and design an adaptive calibration function specifically for predicate classification. This function enables us to dynamically adjust the classification scores for each predicate category. Additionally, we introduce a Distribution Alignment technique that effectively balances the class distribution after the feature extraction phase reaches a stable state, thereby facilitating the retraining of the classification head. Importantly, our Distribution Alignment strategy is model-independent and does not require additional supervision, making it applicable to a wide range of SGG models. Using the scene graph diagnostic toolkit on Visual Genome and several popular models, we achieved significant improvements over the previous state-of-the-art methods with our model. Compared to the TDE model, our model improved mR@100 by 70.5% for PredCls, by 84.0% for SGCls, and by 97.6% for SGDet tasks.

• Journal of Industrial Convergence
• /
• v.21 no.12
• /
• pp.45-54
• /
• 2023

This study aims to examine trends so far and proactively seek future policy directions because the degree of implementation of hyper-connected society and smart technology at home and abroad cannot be overlooked at present in Korea. The method of the study focused on differentiating it from this study through literature research and comparison of major previous studies. As a result of the study, data security and maintenance, enhanced privacy of users and users, and related policy directions for entering a super-aged society were identified in the era of the 4th industrial revolution in Korea according to hyper-connected society and smart technology. Research limitations were difficulties in obtaining data and technical limitations in statistical quantification through trend analysis, although research analysis should be approached through quantitative and quantitative methods. It needs to be supplemented in future studies.

• Journal of TMJ Balancing Medicine
• /
• v.13 no.sup
• /
• pp.9-14
• /
• 2023

Objectives: The purpose of this study was to investigate patients' demand for music and art therapy in the treatment of Parkinson's disease and to understand how patients view the use of music and art therapy as a complementary form of treatment that can improve their quality of life and increase their self-efficacy in addition to medication. Methods: To this end, 60 Parkinson's patients visiting Daejeon Korean Medicine Hospital at Daejeon University were surveyed about music and art therapy. Results: The survey demonstrated a highly positive response to the music program, with strong belief in its efficacy for Parkinson's disease and significant interest in participation. Group music therapy was perceived as effective for enhancing mental health, respiratory function, language disorders, and overall quality of life. Although the art program was also deemed effective, patients showed slightly lower interest and willingness to participate compared to music therapy. The patients preferred 60-minute sessions once a week for about 8 weeks. Conclusions: We propose the following directions for a Parkinson's music and art therapy program. First, develop a once-a-week, 8-week music and art program for Parkinson's patients. Second, design the music program to focus on improving respiratory function and quality of life. Third, the art program focuses on creating a sense of security and comfort through emotional exchange among patients and improving self-efficacy.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.6
• /
• pp.131-141
• /
• 2024

The aging population and worsening lifestyle habits have increased the risk of chronic diseases. This has heightened the importance of preventive healthcare, particularly through personalized health management services based on individual health data. Despite this, the domestic digital healthcare industry remains underdeveloped. Given the need for acceptance from both consumers and providers, this study uses the Analytic Hierarchy Process (AHP) to identify success factors for health management service platforms. AHP evaluates the relative importance of various factors to aid decision-making. Results show that providers prioritize data analysis and platform design, laws and regulations, and data standardization, while consumers prioritize system stability, laws and regulations, and system security. These findings highlight the need for strategies to bridge the expectation gap to effectively promote health management service platforms.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.111-119
• /
• 2024

In this paper, we propose a big data platform that can collect information from various sources collected at ocean. Currently operating ocean-related big data platforms are focused on storing and sharing created data, and each data provider is responsible for data collection and preprocessing. There are high costs and inefficiencies in collecting and integrating data in a marine environment using communication networks that are poor compared to those on land, making it difficult to implement related infrastructure. In particular, in fields that require real-time data collection and analysis, such as weather information, radar and sensor data, a number of issues must be considered compared to land-based systems, such as data security, characteristics of organizations and ships, and data collection costs, in addition to communication network issues. First, this paper defines these problems and presents solutions. In order to design a big data platform that reflects this, we first propose a data source, hierarchical MEC, and data flow structure, and then present an overall platform structure that integrates them all.

• Journal of Environmental Science International
• /
• v.33 no.6
• /
• pp.417-425
• /
• 2024

This study investigated the effects of various planting models on the joint cultivation of eggplant (Solanum melongena) and marigold (Tagetes erecta)to enhance sustainable rooftop urban farming. Rooftop agriculture is increasingly valued to boost the food supply and benefit the environment. Integrating such practices into urban planning is viewed as a way to sustainably manage resources and improve the food-energy-water cycle in cities. The experiment was conducted on a rooftop in Chungju, South Korea from May to August. Four different planting setups were used: central eggplant with peripheral marigold (SET), eggplant with a protective net (SIC), central marigold with peripheral eggplant (TES), and control with only eggplant (CON S). These models tested the effects of companion planting versus monoculture using a lightweight soil mix ideal for rooftops made from cocopeat and perlite and enriched with organic fertilizer. Measurements focused on soil conditions and plant health and assessed soil temperature, moisture, conductivity, plant height, width, and leaf size. The results indicated that the SET modelyielded the best growth. This setup benefited from marigold pest control properties and its ability to improve soil conditions by enhancing moisture and nutrient levels and aiding eggplant growth. These findings underscore the potential of mixed planting on rooftops and suggest that such approaches can be effectively incorporated into urban agriculture to boost yield and environmental sustainability. This study supports the idea that diverse planting methods can significantly affect plant growth and promote urban greening and food security.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.165-172
• /
• 2024

In this paper, we designed a smart home system that can be used intuitively and easily in everyday life, such as sending text messages to users, providing various information and scheduling using smart AI, and providing lighting and atmosphere suitable for the atmosphere in situations such as listening to music using neopixels, as well as using ESP32, RFID, and Google Cloude Console using raspberry pie. As a result of the experiment, it was confirmed that security characters were normally sent to users when RFID was recognized on ESP32 connected to Wi-Fi even if the power was reconnected, and smart AI using Neopixel lighting, Raspberry Pie, and voice recognition, which calculated frequency, also changed the recognition rate over distance, but it worked.