이경순 Kyung-soon Lee , 조용운 Yong-un Cho , 김민정 Min-jung Kim

DOI: JANT Vol.28(No.5) 582-587, 2024

---

The azimuth generator and acquirer for UAV navigation is a system that helps UAVs to precisely determine their own azimuth while they are flying, similar to the DVOR of existing civil aircraft. This system consists of a ground base station that generates an azimuth signal by the Doppler effect and a signal acquisition terminal that receives the azimuth signal and derives the azimuth of the UAV. This study focuses on the implementation of C-band radio navigation in the 5.03 ~ 5.15 GHz terrestrial band to cooperate with GNSS navigation mainly used in existing UAMs. This is one of the navigation technologies that can fully satisfy the requirements of Title 14 of CFR-135.165. According to the FAA, the use of two or more independent navigation sources for aircraft is proposed for aircraft. In this study, the design and implementation of the core technologies of the system, such as azimuth signal generation, sine/cosine signal generation required for azimuth derivation, phase detection comparator of AM/FM signals, and blending signal generator for antenna switching for the Doppler effect, were performed using FPGA. The results of azimuth derivation for 1000 samplings were confirmed to satisfy the ICAO standard of ±3° with a maximum deviation of ±1.1°.

정환호 Hwanho Jeong , 박건환 Geonhwan Park , 박소라 So-ra Park , 손민혁 Minhyuk Son

DOI: JANT Vol.28(No.5) 588-593, 2024

---

Korea augmentation satellite system (KASS), KASS training and test platform (KTTP) was developed for KASS operator and maintainer training and KASS system verification, and it is an independently operated system. KTTP has one each of KASS reference station (KRS), KASS processing station (KPS), KASS uplink station (KUS), network bench, geostationary earth orbit (GEO) simulator and added the one central monitoring & control simulator (CMS) to switch the master operation mode to backup. Hardware and software of KTTP are identical to the KASS system. Therefore, the area where KTTP is installed must be verified through site infrastructure acceptance review (SIAR) in the same way as the KASS system. KTTP Verification is based on the KASS system integration, verification, qualification (IVQ) procedure, and verification was completed with a focus on the functionality rather than performance.

조현수 Hyunsoo Cho , 김은정 Eunjung Kim , 김성엽 Sung-yeob Kim , 이명식 Myeongsik Lee

DOI: JANT Vol.28(No.5) 594-600, 2024

---

With the operation of the Korean satellite-based augmentation System (KASS), it has become possible to develop and implement localizer performance with vertical guidance (LPV) instrument approach procedures. In this study, we developed an LPV procedure for Runway 18 at Ulsan Airport, where the installation of an instrument landing system (ILS) is unable due to obstacles, and examined the potential reduction in flight cancellations. The study results suggest that, when developing an LPV instrument approach procedure, the decision altitude would be 380 feet, and the visibility minimum would be 1,200 meters. Consequently, it is expected that flight cancellations for Runway 18, based on 2023 cancellation data, could be reduced by 92% under the given visibility and ceiling conditions if LPV instrument approach procedure is developed. Therefore, the development and operation of the LPV instrument approach procedure for Runway 18 at Ulsan Airport is deemed essential.

김홍석 Hong-suk Kim , 반규리 Gyu-ri Ban , 서지훈 Ji-hun Seo , 정윤호 Yunho Jung

DOI: JANT Vol.28(No.5) 601-609, 2024

---

In this paper, we present the design and implementation results of a system that classifies drones from other objects using an FMCW (frequency-modulated continuous wave) radar sensor. The proposed system detects various objects through a four-stage signal processing procedure, consisting of FFT, CFAR, clustering, and tracking, using signals received from the radar sensor. Subsequently, a deep learning process is conducted to classify the detected objects as either drones or other objects. To mitigate the high computational demands and extensive memory requirements of deep learning, a BNN (binary neural network) structure was applied, binarizing the CNN (convolutional neural network) operations. The performance evaluation and verification results demonstrated a drone classification accuracy of 89.33%, with a total execution time of 4 ms, confirming the feasibility of real-time operation.

최지헌 Jihun Choi

DOI: JANT Vol.28(No.5) 610-615, 2024

---

Based on the K-UAM roadmap, this study evaluated the safety for the initial UAM demonstration project and operation. Research in each field is underway for early settlement of UAM, but the progress of research is showing limitations without clear regulations and lack of operational experience. This study aims to find a way for UAM to operate in reality, and for this purpose, the defense system for evaluating the safety of civil aviation was applied as SMS. In addition, the technical elements of the maneuvering of UAM aircraft were reviewed from a flight perspective. As a result, a unexpected problem was derived in the take-off and landing stage of the current airspace, flight corridor, and operation stage. Based on this, alternatives were derived and suggested to successfully complete the UAM demonstration project by FOSA which is Risk assessment Tool. Therefore, if the direction of UAM research is set by citing the results of this study, it is expected to contribute to the safe operation of UAM.

조인철 In-cheol Cho , 조성훈 Sung-hoon Cho , 안지훈 Ji-hoon An , 문현욱 Hyun-wook Moon , 손성환 Sung-hwan Sohn , 정태희 Taehee Jeong , 임상호 Sangho Lim

DOI: JANT Vol.28(No.5) 616-622, 2024

---

Recently, each country is competitively investing huge amounts of money in space development, and space risks such as tens of thousands of satellites, spy satellites, and space debris are increasing. Although there are several assets for monitoring space, interest is growing in the detection of space objects using radar, which is useful for constant alert and wide-area surveillance. While the existing radar system was operated to detect aircraft, ships, and missiles within hundreds of kilometers, radar to detect space objects must be expanded to thousands of kilometers. In this case, the size of data that impairs signal processing increases, which can lead to a decrease in radar detection performance. In this paper, we present a method of speeding up signal processing using GPU when detecting space objects, which are long-distance targets, and the results of comparing performance compared to existing CPUs.

김하은 Ha-eun Kim , 김상일 Sang-il Kim , 이성섭 Soung-sub Lee

DOI: JANT Vol.28(No.5) 623-631, 2024

---

Currently, the Korean military is promoting a small satellite system project to establish a command and control system that can quickly respond to various threats from all directions, not only from North Korea. In this paper, we present a nonlinear control law to acquire the geometry of a small cluster satellite deployment using a low-thrust electric propulsion system, the Hall thruster. The control law is obtained by applying the general Lyapunov’s control theory and analytical solution for the on-off control of the electric thruster, and it is a technique that can effectively deploy the orbital configuration of the satellite constellation by controlling only the semi-major axis of the six elements of the orbit. In this study, the stability of the developed control law is verified and the results are analyzed for application to a very small satellite system.

정성재 Seong-jae Jeong , 정다나 Da-na Jung , 조성희 Sung-hee Cho , 김민선 Min-seon Kim , 남덕우 Deok-woo Nam

DOI: JANT Vol.28(No.5) 632-639, 2024

---

In this paper, the analysis method and results of audio data through LVDS interface and data Acquisition of the intercom. Intercom uses LVDS interface to transmit/receive digital audio data between audio interface unit and intercom control panel and to data acquisition of audio signals. Data acquisition using LVDS interface is essential for downloading various audio signals interfaced to the intercom to a PC and performing data analysis on noise, audio qualitys, etc. To achieve this, hardware for audio data acquisition was configured in the intercom test equipment. As a result, individual analysis of intercom input/output audio signals before and after logic processing was possible, which was effective in deriving the cause of noise signals. In addition, by data acquisition on actual voice communication and audio signals in ac system integrated laboratory and an environment/ electromagnetic test, audio signal level measurements and frequency analysis were performed to debug noise signal removal and verify the audio quality of the intercom.

유태훈 Tae-hoon Yoo , 방현태 Hyeon-tae Bang , 윤원근 Won-keun Youn

DOI: JANT Vol.28(No.5) 640-649, 2024

---

In this study, we introduce an interactive multi-model-probabilistic data association filter (IMM-PDAF), a multi-target tracking algorithm that integrates multiple dynamic models for accurate real-time maritime target tracking. Multi-target tracking in the maritime environment requires high accuracy due to the complex dynamic environment and various movement patterns. The existing CV-PDAF (constant velocity model) and CT-PDAF (circling model) each assume a constant movement pattern, but it is difficult to handle all the complex movements occurring in various maritime environments with these single models. To solve this problem, this study proposes an interactive multi-model-probabilistic data association filter (IMM-PDAF), and the results of this paper applied to maritime RADAR data show that the proposed IMM-PDAF has relatively lower RMSE values than CV-PDAF and CT-PDAF, and has strong positioning performance even in complex dynamic environments. Therefore, this study results highlight the potential of the proposed IMM-PDAF to improve the reliability and efficiency of maritime surveillance systems and provide a multi-target tracking solution for complex tracking environments.

정상규 Sang Gyu Jeong , 권나은 Na Eun Kwon , 김형우 Hyung Woo Kim

DOI: JANT Vol.28(No.5) 650-656, 2024

---

This study proposes an AI-based image restoration technique to reduce image distortion caused by lighting and noise in nighttime environments and improve the performance of infrared detection systems. A synthetic image dataset was constructed using visible light images under various lighting conditions and ISO settings, and deep learning models (AutoEncoder and U-Net) were trained to assess image restoration performance. Experimental results show that the Multi-ISO model (9-channel) outperforms the Single-ISO model (3-channel), especially when utilizing input data with multiple ISO values. This study demonstrates that AI models can be effectively trained using synthetic data, even when real data collection is challenging, and can be applied to image restoration tasks. These findings are expected to contribute to enhancing the performance of optical detection systems through AI-based technology.