• 한국음향학회지
• /
• 제43권1호
• /
• pp.78-88
• /
• 2024

수중 소음 측정이 가능한 수동 소나에 수신된 선박 방사소음은 Detection of Envelope Modulation on Noise(DEMON) 분석으로 얻은 선박 정보를 사용하여 선박 식별과 분류가 가능하다. 하지만 낮은 신호대잡음비(Signal-to-Noise Ratio, SNR) 환경에서는 DEMON 그램 내 선박 정보가 담겨있는 표적 주파수선을 분석 및 파악하는데 어려움이 발생한다. 본 논문에서는 낮은 SNR 환경에서 보다 정확한 표적 식별을 위해 딥러닝 기법 중 의미론적 분할을 사용하여 표적 주파수선들을 추출하는 연구를 수행하였다. SNR과 기본 주파수를 변경시키며 생성한 모의 DEMON 그램 데이터를 사용하여 의미론적 분할 모델인 U-Net, UNet++, DeepLabv3+를 학습 후 평가하였고, 학습된 모델들을 이용하여 캐나다 조지아 해협에서 측정한 선박 방사소음 데이터셋인 DeepShip으로 제작한 DEMON 그램 예측 성능을 비교하였다. 모의 DEMON 그램으로 학습된 모델을 평가한 결과 U-Net이 성능이 가장 높았으며, DeepShip으로 만든 DEMON 그램의 표적 주파수선을 어느 정도 추출할 수 있는 것을 확인하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.1062-1073
• /
• 2007

Acoustic Target Strength (TS) is a major parameter of the active sonar equation, which indicates the ratio of the radiated intensity from the source to the re-radiated intensity by a target. In developing a TS equation, it is assumed that the radiated pressure is known and the re-radiated intensity is unknown. This research provides a TS equation for polygonal plates, which is applicable to near field acoustics. In this research, Helmholtz-Kirchhoff formula is used as the primary equation for solving the re-radiated pressure field; the primary equation contains a surface (double) integral representation. The double integral representation can be reduced to a closed form, which involves only a line (single) integral representation of the boundary of the surface area by applying Stoke's theorem. Use of such line integral representations can reduce the cost of numerical calculation. Also Kirchhoff approximation is used to solve the surface values such as pressure and particle velocity. Finally, a generalized definition of Sonar Cross Section (SCS) that is applicable to near field is suggested. The TS equation for polygonal plates in near field is developed using the three prescribed statements; the redection to line integral representation, Kirchhoff approximation and a generalized definition of SCS. The equation developed in this research is applicable to near field, and therefore, no approximations are allowed except the Kirchhoff approximation. However, examinations with various types of models for reliability show that the equation has good performance in its applications. To analyze a general shape of model, a submarine type model was selected and successfully analyzed.

• 한국소음진동공학회논문집
• /
• 제15권11호
• /
• pp.1303-1310
• /
• 2005

Stabilization error of target aiming systems mounted on moving vehicles is an important performance because the error directly affects hit Probability. In a heavy load targetaiming system, the disturbance torque comes from mass unbalance and linear acceleration is a main source of stabilization error. This study suggests an experimental design method of disturbance feedforward compensation control to improve the stabilization performance of heavy load target aiming systems. The filtered_x least square(FxLMS) algorithm is used to estimate the compensator coefficients adaptively. The proposed control is applied to a simple experimental set-up which simulates dynamic characteristics of a real target aiming system. The feasibility of the proposedtechnique is illustrated, along with results of experiments.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.635-638
• /
• 1998

This paper proposes a simple method to measure system's performance in target tracking problems. Essentially employing the Cramer-Rao lower bound (CRLB) on trakcing accuracy, an algorithm of predicting system's performance under various scenarios is developed. The input data is a collection of measurements over time fromsensors embedded in gaussian noise. The target of interest may not maneuver over the processing time interval while the own ship observing platform may maneuver in an arbitrary fashion. Th eproposed approach is demonstrated and discussed through simulation results.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1988년도 한국자동제어학술회의논문집(국내학술편); 한국전력공사연수원, 서울; 21-22 Oct. 1988
• /
• pp.183-186
• /
• 1988

The problem addressed in this paper is the accurate tracting of a dynamic target using outputs from a forward - looking infrared(FLIR) sensor as measurements. The important variations are 1) the spread of the target intensity pattern in the FLIR image plane, 2) target motion characteristics, and 3) the rms value and both spartial and temporal correlation of the back - ground noise. Based on this insights. design modifications and on - line adaptation copabilities are incorporated to enable this type of filter track highly maneuverable targets such as air-to-air missiles, with spatially distributed and changing image intensity profiles, against, background clutter.

• 한국음향학회지
• /
• 제20권6호
• /
• pp.17-25
• /
• 2001

기존의 적응형 잡음제거 기법을 적용하여 잔향음의 영향을 제거하는 경우, 수렴속도, 추적기능, 수치적 안전성, 표적신호와 잔향음 신호간의 상대적인 주파수 특성, 그리고 표적속도 등의 상호 배반적인 요인에 의하여 그 성능이 달라진다. 본 논문에서는 수렴속도, 연산량 그리고 검출능력 등의 요구조건들을 상호 보완하는 적응형 잔향음 제거 기법을 제안하였다. 본 논문에서 제안한 기법은 변환영역에서 동작하도록 설계하였으며, Gram-schmidt 직교화 과정을 적용하였고, 이를 에스컬레이터 알고리즘으로 구현하였다. 먼저 변환영역 알고리즘을 사용함으로써 수렴속도와 수치적 안전성 혹은 연산량 측면에서의 성능향상을 유도할 수 있다. 제안된 잔향음 제거 기법을 삼척시 연안에서 실측한 자료에 적용하여 그 성능을 검증하였다. 모의실험을 통하여 제안된 기법이 잔향음 신호에 근접한 주파수 대역의 표적신호를 보존하고, 정지표적과 이동표적에 대해서도 성능의 변화가 상대적으로 적으면서 효율적으로 잔향음을 제거한다는 사실을 확인하였다.

• 한국정보전자통신기술학회논문지
• /
• 제12권4호
• /
• pp.434-439
• /
• 2019

본 논문에서는 신호 대 잡음비를 향상시키기 위해서 시간지연을 보상하는 제어 알고리즘을제안하고, 제안한 제어 알고리즘을 연속파 레이더 방정식에 적용하여 목표물의 정보를 추정한다. 제안한 제어 알고리즘은 신호 대 잡음비를 향상시키기 위해서 수신신호에 가중치를 곱하여 각 배열소자의 출력신호를 향상시킨다. 레이더는 공간상에 신호를 방사하고 불체로부터 반사된 신호를 수신하여 정보를 추출한다. 그러나 무선통신환경에서는 인공 및 자연구조물로 인해서 다중 경로현상이 발생되어 시호의 시간지연이 나타나게 된다. 시간지연은 간섭신호 및 신호왜곡 현상으로 시스템 성능이 저하되어 원하는 정보를 정확히 추정하기 어렵다. 본 연구에서는 등 간격 선형 배열 안테나를 사용하여 수신신호에 가중치로 지연신호를 보상함으로서 목표물의 추정 정보를 향상 시킬 수 있다. 모의실험으로 본 연구에서 제안한 제어 알고리즘과 지연시간이 보상되지 않은 방법에 대해서 성능을 비교 분석하였으며 본 연구에서 제안한 제어 알고리즘이 원하는 목표물 거리추정 정보가 향상되었음을 입증하였다.

• Journal of Audiology & Otology
• /
• 제25권4호
• /
• pp.171-177
• /
• 2021

Background and Objectives: The purpose of the present study was to validate the performance and diagnostic efficacy of the Korean digits-in-noise (K-DIN) test in comparison to the Korean speech perception-in-noise (K-SPIN) test, which is the representative speech-in-noise test in clinical practice. Subjects and Methods: Twenty-seven subjects (15 normal-hearing and 12 hearing-impaired listeners) participated. The recorded Korean 0-9 digits were used to form quasirandom digit triplets; 50 target digit triplets were presented at the most comfortable level of each subject while presenting speech-shaped background noise at various levels of signal-to-noise ratios (-12.5, -10, -5, or +5 dB). Subjects were then instructed to listen to both target and noise masker unilaterally and bilaterally through a headphone. K-SPIN test was also conducted using the same procedure as the K-DIN. After calculating their percent correct responses, K-DIN and K-SPIN results were compared using a Pearson-correlation test. Results: Results showed a statistically significant correlation between K-DIN and K-SPIN in all hearing conditions (left: r=0.814, p<0.001; right: r=0.788, p<0.001; bilateral: r=0.727, p<0.001). Moreover, the K-DIN test achieved better testing efficacy, shorter average listening time (5 min vs. 30 min), and easier performance of task according to participants' qualitative reports than the K-SPIN test. Conclusions: In this study, the Korean version of digit triplet test was validated in both normal-hearing and hearing-impaired listeners. The findings suggest that the K-DIN test can be used as a simple and time-efficient hearing-in-noise test in audiology clinics in Korea.

• 대한청각학회지
• /
• 제25권4호
• /
• pp.171-177
• /
• 2021

Background and Objectives: The purpose of the present study was to validate the performance and diagnostic efficacy of the Korean digits-in-noise (K-DIN) test in comparison to the Korean speech perception-in-noise (K-SPIN) test, which is the representative speech-in-noise test in clinical practice. Subjects and Methods: Twenty-seven subjects (15 normal-hearing and 12 hearing-impaired listeners) participated. The recorded Korean 0-9 digits were used to form quasirandom digit triplets; 50 target digit triplets were presented at the most comfortable level of each subject while presenting speech-shaped background noise at various levels of signal-to-noise ratios (-12.5, -10, -5, or +5 dB). Subjects were then instructed to listen to both target and noise masker unilaterally and bilaterally through a headphone. K-SPIN test was also conducted using the same procedure as the K-DIN. After calculating their percent correct responses, K-DIN and K-SPIN results were compared using a Pearson-correlation test. Results: Results showed a statistically significant correlation between K-DIN and K-SPIN in all hearing conditions (left: r=0.814, p<0.001; right: r=0.788, p<0.001; bilateral: r=0.727, p<0.001). Moreover, the K-DIN test achieved better testing efficacy, shorter average listening time (5 min vs. 30 min), and easier performance of task according to participants' qualitative reports than the K-SPIN test. Conclusions: In this study, the Korean version of digit triplet test was validated in both normal-hearing and hearing-impaired listeners. The findings suggest that the K-DIN test can be used as a simple and time-efficient hearing-in-noise test in audiology clinics in Korea.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1999-2005
• /
• 2005

Collaboration in wireless sensor networks must be fault-tolerant due to the harsh environmental conditions in which such networks can be deployed. This paper focuses on finding signal processing algorithms for collaborative target detection based on the generalized approach to signal processing in the presence of noise that are efficient in terms of communication cost, precision, accuracy, and number of faulty sensors tolerable in the wireless sensor network. Two algorithms, namely, value fusion and decision fusion constructed according to the generalized approach to signal processing in the presence of noise, are identified first. When comparing their performance and communication overhead, decision fusion is found to become superior to value fusion as the ratio of faulty sensors to fault free sensors increases. The use of the generalized approach to signal processing in the presence of noise under designing value and decision fusion algorithms in wireless sensor networks allows us to obtain the same performance, but at low values of signal energy, as under the employment of universally adopted signal processing algorithms widely used in practice.