• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.1
• /
• pp.82-95
• /
• 2022

Since military weapon accidents or breakdowns are directly linked to enormous damage, it is important to analyze the causes of weapons system accidents. Recently, in the defense sector, there have been cases in which budget has been saved through analysis of the causes of frequent breakdowns and improvement activities that have occurred in the process of operating weapon systems since 2015. But due to the nature of the defense sector, it is not easy to collect data and studies on weapons system accidents have been insufficient so far. Therefore, this study aims to investigate the causes and types of military weapon accidents by collecting military weapon accident data for military weapon systems and analyzing trends by weapon system classification through the analysis process. It analyzes statistically and visually through social network analysis, NodeXL. It is expected that this study will help improve the stability of the weapon system by reducing the number of military weapon accidents and failures.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.921-930
• /
• 2020

The Counter-battery radar estimates the origin and impact point of the artillery by tracking the trajectory of the shell. In addition, it has the ability of identifying the type of weapon. Depending on the position between the shell and the radar, the detected signals appear differently. This has ambiguity to distinguish the type of shells. This paper compares fuzzy logic and artificial intelligence, which classifies type of shell using the parameter of signal processing step. According to the research result, artificial intelligence can improve identification rate of type of shell. The data used in the experiment was obtained from a live fire detection test.

• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.1
• /
• pp.46-57
• /
• 2022

In this paper, we proposed the DFMEA Implementation Method for safety design of Weapon System. First, we presented the process for DFMEA. And then, the case analysis of OOO missile was performed in accordance with the process presented. After defining the system requirements of OOO missile, failure definition scoring criteria was set. In order to clarify the definition of failure, the failure was classified into safety, reliability, maintainability and others. After performing the function analysis, the relationship matrix analysis was performed to identify the failure mode according to the function without omission. After clarifying the failure classification, mode of failure, cause of failure and effect were analyzed to calculate the severity, occurrence and detection values. After the action priority was judged, the recommended action according to the failure classification was identified for the determined action priority. The results of this study can be used as a relevant basis for the design reflection and resource re-allocation of stakeholders.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.673-687
• /
• 2023

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.

• Journal of Software Engineering Society
• /
• v.29 no.1
• /
• pp.7-12
• /
• 2020

As interest in functional safety has recently increased, application of functional safety standards has been required in various industrial fields. A functional safety standard is a document that defines functional safety-related activities required to prevent system malfunctions. All activities defined in this standard are required differentially according to the classification results calculated through the risk analysis and assessment of the system. In the field of domestic weapon systems, there is a manual for the development and management of weapon system software issued by the Defense Acquisition Program Administration (DAPA ). This manual requires static and dynamic analysis of software for functional safety related activities. However, the manual does not specifically address the classification activity through risk analysis and assessment as required for the preceding activities. Therefore, in this study, we analyze the problems of the manual based on the representative functional safety standards, and propose improvement plans.

• Journal of the Korea Society for Simulation
• /
• v.28 no.3
• /
• pp.11-23
• /
• 2019

An AddSIM(Adaptive distributed and parallel Simulation environment for Interoperable and reusable Models) is developed to construct a composite environment that can be used in the overall stage from military demand analysis to test and evaluation. In addition, a base system model(BSM), which is a component model of the weapon system with standardized hierarchies, has been developed. This paper describes the critical design of BSM for the guided weapon system that can be operated in AddSIM. The guided weapon system BSM is designed for reusability and interoperability, and to have the same interface for assembly, even if the subcomponents have different resolution. Then, each subcomponent is defined and implemented according to the component resolution classification scheme. Finally, Combinations of subcomponents have been used to construct the guided weapon system of various resolution and the performance is compared and analyzed through simulation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.4
• /
• pp.457-465
• /
• 2021

This study analyzed the reality and problems of the defense standard classification system. This paper proposes a plan to perform standard management tasks efficiently through case analysis of domestic and foreign standard classification systems. To continuously solve the problem of military product quality in defense technical data, it is necessary to promptly reflect the civilian's excellent technology and benchmark the civilian standard system to manage high-quality defense standards. First, to analyze the reality, the NATO classification system was analyzed through the private KS of domestic and ICS codes, the US defense standard system of overseas. In the case of the Korean military, the reality of the defense standard classification system was grasped through the National Defense Standards Comprehensive System operated by the Defense Acquisition Program Administration. The classification of the ministry of defense's weapon system and force support system is the most suitable classification system for the Korean military, which is classified into eight weapon systems and six force support standard systems for all steps. Specifically, it was classified into 12 major categories, 66 categories, and 352 sub-categories. In this study, the establishment of the defense standard management system can improve the classification system of new defense standards by reflecting the superior technology of the private sector.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.44-56
• /
• 2018

Researches on military weapons are actively studied to improve national defense power of each country. The military weapon system is being used not only as a weapon but also as a reconnaissance and surveillance device for places where it is difficult for people to access. If such a weapon system becomes an object of attack, military data that is important to national security can be leaked. Furthermore, if a device is taken, it can be used as a terrorist tool to threaten its own country. So, security of military devices is necessarily required. In order to enhance the security of a weapon system such as drone, it is necessary to form a chain of trust(CoT) that gives trustworthiness to the overall process of the system from the power on until application is executed. In this paper, by analyzing the trusted computing-based boot technology, we derive trusted boot technology components and classify them based on hardware dependence/independence. We expect our classification of hardware dependence/independence to be applied to the trusted boot technology of our self-development ultraprecision weapon system to improve the defense capability in our military.

• Structural Engineering and Mechanics
• /
• v.17 no.3_4
• /
• pp.409-428
• /
• 2004

In this paper, a Holder exponent, a measure of the degree to which a signal is differentiable, is presented to detect the presence of a discontinuity and when the discontinuity occurs in a dynamic signal. This discontinuity detection has potential applications to structural health monitoring because discontinuities are often introduced into dynamic response data as a result of certain types of damage. Wavelet transforms are incorporated with the Holder exponent to capture the time varying nature of discontinuities, and a classification procedure is developed to quantify when changes in the Holder exponent are significant. The proposed Holder exponent analysis is applied to various experimental signals to reveal underlying damage causing events from the signals. Signals being analyzed include acceleration response of a mechanical system with a rattling internal part, acceleration signals of a three-story building model with a loosing bolt, and strain records of an in-situ bridge during construction. The experimental results presented in this paper demonstrate that the Holder exponent can be an effective tool for identifying certain types of events that introduce discontinuities into the measured dynamic response data.

• Journal of the Society of Disaster Information
• /
• v.11 no.3
• /
• pp.400-405
• /
• 2015

Operating system of CBRNE(Chemical, Biological, Radiological, Nuclear, and High-yield explosive)) weapon in North Korea has reached an internationally significant level. In preparation against the CBRNE weapon attack, the US is securing various forms of defense shelters and operating it based on classification by disaster characteristics. However, it is currently difficult to expect an efficient protective ability from South Korea due to the reckless designation of defense shelters without consideration of disaster characteristics. At this, this study examined the present condition of formerly used facilities, analyzed the characteristics of each facilities for the sorting of defense shelters that are possible of conversion into shelter against CBRNE weapon, and presented results and proposals gained through this study research.