• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.4
• /
• pp.352-360
• /
• 2023

Recently, AI pilots using reinforcement learning are developing to a level that is more flexible than rule-based methods and can replace human pilots. In this paper, a curriculum was used to help head-on combat with reinforcement learning. It is not easy to learn head-on with a reinforcement learning method without a curriculum, but in this paper, through the two circle-based head-on air combat learning technique, ownship gradually increase the difficulty and become good at head-on combat. On the two-circle, the ATA angle between the ownship and target gradually increased and the AA angle gradually decreased while learning was conducted. By performing reinforcement learning with and w/o curriculum, it was engaged with the rule-based model. And as the win ratio of the curriculum based model increased to close to 100 %, it was confirmed that the performance was superior.

• Journal of the Korean Society of Systems Engineering
• /
• v.20 no.1
• /
• pp.64-75
• /
• 2024

With the development of the Internet, the information in our lives has become fast and diverse. Especially in the field of defense, articles and information are pouring in from various sources every day, and fast information selection, understanding, and decision-making are required in the ever-changing situation. It is very cumbersome to go from platform to platform and read articles one by one to get the information you need. To solve this problem, this research aims to save time and provide quick access to the latest information by allowing you to quickly grasp key information from summarized content without having to read the entire article. This can improve efficiency by allowing defense professionals to focus more on important tasks rather than extensive information search and analysis.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.2
• /
• pp.256-264
• /
• 2024

In the last 10 years, AI has made rapid progress, and image classification, in particular, are showing excellent performance based on deep learning. Nevertheless, due to the nature of deep learning represented by a black box, it is difficult to actually use it in critical decision-making situations such as national defense, autonomous driving, medical care, and finance due to the lack of explainability of judgement results. In order to overcome these limitations, in this study, a model description algorithm capable of local interpretation was applied to the inception network-derived AI to analyze what grounds they made when classifying national defense data. Specifically, we conduct a comparative analysis of explainability based on confidence values by performing LIME analysis from the Inception v2_resnet model and verify the similarity between human interpretations and LIME explanations. Furthermore, by comparing the LIME explanation results through the Top1 output results for Inception v3, Inception v2_resnet, and Xception models, we confirm the feasibility of comparing the efficiency and availability of deep learning networks using XAI.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.5
• /
• pp.549-557
• /
• 2013

This paper describes a block processing approach to detect targets in front of mono-static terrain imaging radar (TIR). It is difficult to employ several conventional imaging methods of the synthetic aperture radar(SAR) because the TIR is an ultra-wide-band(UWB) type of radar and employs a dechirp-on-receive process. To design an available imaging method, a block processing approach which conducts a range compression and an azimuth compression is proposed in this paper. The complete derivation of the proposed approach is presented. The results of simulations and field tests are demonstrated to show the performance and validity of the proposed approach.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.2
• /
• pp.181-186
• /
• 2017

Synthetic aperture radar(SAR) has been widely used for reconnaissance. It provides high-resolution, day-and-night and weather-independent images for a multitude of applications. Because SAR coherently combines many viewing angles to effectively create a large aperture(narrow beam) radar, the test-bed should be capable of moving straightly SAR sensor for the integration angle to meet resolution. This paper describes the test-bed developed to test and evaluate the SAR performance. It forms high-quality images in real time and saves the raw data for the purpose of post processing on the ground.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.1
• /
• pp.201-210
• /
• 2011

Each area of society has changed because of the development of information technology. Especially, the advent of NCW based on the technology of network has become a new paradigm for executing warfare. Effectiveness of NCW can be maximized by building the C4I system which is a core system of NCW. However, if we don't consider the influence in term of human dimension, we can't expect the effect of C4I system, since the key factor in C4I is human. In this paper, we propose an algorithm for evaluating Combat Power Effectiveness by considering the Influence of Human Factors that wasn't reflected in the past. Based on experimental validation our algorithm is more substantial than baseline algorithms. In addition, we proved that the Influence of Human Factors(e.g. collaboration) is the most important in battlefield. Therefore, proposed algorithm can be used for enhancing not only mission effectiveness in terms of military field but also work performance by effective Human Resource Management in terms of an enterprise.

• Journal of the Korea Society for Simulation
• /
• v.15 no.4
• /
• pp.69-77
• /
• 2006

In most positioning and allocation practices, many mathematical models are proposed in various fields. The set covering (SC) problem has many practical applications of modeling not only real world problem but also in military. As our air defense weapon systems are getting older and declining the performance, new plans far acquisition of high-tech air defense weapon system are being conducted. In this paper we established simulation model for optimal allocation of KDX which carries new missile defense weapon system by using partial set covering considering both attacker and defender side. By implementating simulation model, we assess the available scenarios and show the optimal pre-positioning of KDX and interceptor's allocation. Furthermore, we provide a variety of experiments and extensive scale sized situations for Korea Indigenous Missile Defense (KIMD) and support decision-making for efficient positioning of unit.

• Journal of Digital Convergence
• /
• v.18 no.1
• /
• pp.1-10
• /
• 2020

In this study, an empirical analysis using Difference-in-Difference(DID) methods was conducted based on the data of domestic defense companies to look at the impact of government support on the performance of these defense industries. As a result, we found that the volume of sales and assets of defense enterprises does not increase or decrease solely on the benefit of government support. In addition, government support for net profit to total sales ratio and net profit to total assets ratio improved significantly, but the effect was only seen in the short term. Unlike previous studies that focused on identifying the existence of differences between industries, our research is different in that it is the first to analyze empirically within the Korean defense industry. It is also theoretically meaningful in that it looked at the effects of government support, especially in industries with unique characteristics that have yet to be known.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.1
• /
• pp.66-74
• /
• 2015

Conventional hydrophones can only measure acoustic pressure. To measure both acoustic pressure and incident direction, various types of vector hydrophones have been researched. In this paper, we deal with a ring-type multimode vector hydrophone divided into 4 elements and present a direction-of-arrival (DoA) estimation method based on the pressure gradient-acceleration relationship. The performance of the presented method is analyzed by the simulation based on the sensor modeling and is verified by the water tank experiment. The proposed method could work under the multi-frequency condition and may be utilized in many applications due to its low computation complexity.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.5
• /
• pp.871-879
• /
• 2020

Perceptual Ad-Blocking is a new advertising blocking technique that detects online advertising by using an artificial intelligence-based advertising image classification model. A recent study has shown that these Perceptual Ad-Blocking models are vulnerable to adversarial attacks using adversarial examples to add noise to images that cause them to be misclassified. In this paper, we prove that existing perceptual Ad-Blocking technique has a weakness for several adversarial example and that Defense-GAN and MagNet who performed well for MNIST dataset and CIFAR-10 dataset are good to advertising dataset. Through this, using Defense-GAN and MagNet techniques, it presents a robust new advertising image classification model for adversarial attacks. According to the results of experiments using various existing adversarial attack techniques, the techniques proposed in this paper were able to secure the accuracy and performance through the robust image classification techniques, and furthermore, they were able to defend a certain level against white-box attacks by attackers who knew the details of defense techniques.