• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.3
• /
• pp.1-10
• /
• 2020

In this paper, we propose an algorithm for estimating blood pressure using ECG (Electrocardiogram) and PPG (Photoplethysmography) signals. To estimate the BP (Blood pressure), we generate a periodic input signal, remove the noise according to the differential and threshold methods, and then estimate the systolic and diastolic blood pressures based on the convolutional neural network. We used 49 patient data of 3.1GB in the MIMIC database. As a result, it was found that the prediction error (RMSE) of systolic BP was 5.80mmHg, and the prediction error of diastolic BP was 2.78mmHg. This result confirms that the performance of class A is satisfied with the existing BP monitor evaluation method proposed by the British High Blood Pressure Association.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.4
• /
• pp.501-507
• /
• 2020

This paper proposes a new frame synchronization technique based on convolutional neural network (CNN). The conventional frame synchronizers usually find the matching instance through correlation between the received signal and the preamble. The proposed method converts the 1-dimensional correlator ouput into a 2-dimensional matrix. The 2-dimensional matrix is input to a convolutional neural network, and the convolutional neural network finds the frame arrival time. Specifically, in additive white gaussian noise (AWGN) environments, the received signals are generated with random arrival times and they are used for training data of the CNN. Through computer simulation, the false detection probabilities in various signal-to-noise ratios are investigated and compared between the proposed CNN-based technique and the conventional one. According to the results, the proposed technique shows 2dB better performance than the conventional method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.3C
• /
• pp.248-256
• /
• 2003

Space-time coding was designed for an efficient transmit diversity technique to improve performance of wireless communication. For the transmit diversity using space-time coding, the receiver requires to estimate channel parameters corresponding to each transmit antennas. In this paper, we propose an efficient channel estimation scheme based on trigonometric polynomial approximation in OFDM systems with transmit diversity using space-time coding. The proposed scheme is more efficient than the conventional scheme in terms of the computational complexity. For QAM modulation, when the size of FFH is 128, the conventional scheme with significant tap caching of 7 requires 9852 complex multiplications for TU, HT and BU channels. But the proposed scheme requires 2560, 7680 and 3584 complex multiplications for TU, HT and BU channels, respectively. Especially, for channels with smaller Doppler frequency and delay spreads, the proposed scheme has the improved BER performance and complexity. In addition, we evaluate the performance of maximum delay spread estimation in unknown channel. The performance of the proposed scheme is investigated by computer simulation in various multi-path fading environments.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.17-25
• /
• 2023

In this paper, we propose a model that can perform human pose estimation through a MobileViT-based model with fewer parameters and faster estimation. The based model demonstrates lightweight performance through a structure that combines features of convolutional neural networks with features of Vision Transformer. Transformer, which is a major mechanism in this study, has become more influential as its based models perform better than convolutional neural network-based models in the field of computer vision. Similarly, in the field of human pose estimation, Vision Transformer-based ViTPose maintains the best performance in all human pose estimation benchmarks such as COCO, OCHuman, and MPII. However, because Vision Transformer has a heavy model structure with a large number of parameters and requires a relatively large amount of computation, it costs users a lot to train the model. Accordingly, the based model overcame the insufficient Inductive Bias calculation problem, which requires a large amount of computation by Vision Transformer, with Local Representation through a convolutional neural network structure. Finally, the proposed model obtained a mean average precision of 0.694 on the MS COCO benchmark with 3.28 GFLOPs and 9.72 million parameters, which are 1/5 and 1/9 the number compared to ViTPose, respectively.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.1
• /
• pp.29-38
• /
• 2024

The Matched Field Processing (MFP) is an estimation method for a source range and depth based on the prediction of sound propagation. However, as the frequency increases, the prediction inaccuracy of sound propagation increases, making it difficult to estimate the source position. Recently proposed, the Frequency-Difference Matched Field Processing (FD-MFP) is known to be robust even if there is a mismatch by applying a frequency-difference autoproduct extracted from the auto-correlation of a high frequency signal. In this paper, in order to evaluate the performance of the FD-MFP using a horizontal line array, simulations were conducted in the environment of the East Sea of Korea. In the area of Bottom Bounce (BB) and Convergence Zone (CZ) where detection of a sound source is possible at a long range, and the results of localization were analyzed. According to the the FD-MFP simulations of horizontal line array, the accuracy of localization is similar or degraded compared to the conventional MFP due to diffracted field and mismatch of sound speed. There was no clear result from the simulations conforming that the FD-MFP was more robust to mismatch than the conventional MFP.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2003.06a
• /
• pp.19-28
• /
• 2003

Weighting adjustment is a method of improving the efficiency of the estimator by incorporating auxiliary variables at the estimation stage. One commonly used method of weighting adjustment is the poststratification, which is a special case of regression estimation but is relatively feasible in terms of actual implementation. If too many auxiliary variables are used in the poststratification, the bias of the resulting point estimator is no longer negligible and the final weights may have extreme weights. In this study, we propose a method of weight ing adjustment that compromises the efficiency and the bias of the point estimator. A limited simulation study is also presented.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.1-3
• /
• 2010

본 논문에서는 VSB 수신기를 위한 반송파 위상 오차에 독립적인 심벌 타이밍 옵셋 추정 알고리즘을 제안하고자 한다. 심벌 타이밍 옵셋 추정에 대표적인 알고리즘인 가드너 방법은 반송파 위상 옵셋이 포함된 VSB 수신기에서는 타이밍 옵셋을 추정할 수 없다. 본 논문에서는 수신신호의 공액 곱 연산을 통하여 신호의 스펙트럼을 확장하고 반송파 위상 옵셋을 상쇄 하였고, 그 후 가드너 알고리즘을 통하여 인접 스펙트럼 간의 중복부분을 발생시켜, 타이밍 옵셋을 추정하는 방식을 연구하였다. 시뮬레이션 결과, 제안하는 알고리즘은 VSB 수신기에서 반송파 위상 오차에 영향을 받지 않고, 정확하게 타이밍 옵셋을 추정할 수 있는 것으로 나타났다.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.312-318
• /
• 2022

This paper proposes a model which classifies the type of guns and information about sound source location using deep neural network. The proposed classification model is composed of convolutional neural networks (CNN) and long short-term memory (LSTM). For training and test the model, we use the Gunshot Audio Forensic Dataset generated by the project supported by the National Institute of Justice (NIJ). The acoustic signals are transformed to Mel-Spectrogram and they are provided as learning and test data for the proposed model. The model is compared with the control model consisting of convolutional neural networks only. The proposed model shows high accuracy more than 90 %.

• Journal of Biomedical Engineering Research
• /
• v.21 no.5
• /
• pp.449-455
• /
• 2000

순간적으로 승모판에서 혈류가 역류하는 영역을 측정하기 위해서, PISA 방법이 자주 이용되고 있다. 이 방법은 물질보존법칙에 근거하여, 구멍을 통과하는 유체량을 isotach 표면적과 이에 대응하는 속도의 곱으로 구하는 것이다. 이러한 PISA 방법에서 사용되는 유동모델은 반구모델과 비반구모델의 형태인데, 이는 isotach 표면적이 반구이거나 비반구임을 가정하여 계산된 것이다. 이러한 isotach 모델링에서는 isotach의 높이와 폭의 결정이 유체량을 추정하는데 아주 중요한 변수가 된다. 본 연구에서는 in-vitro 칼라 도플러 영상으로부터 PISA 영역을 추정을 위하여 영역기반을 근간으로 하는 비반구모델에 대한 표면적 추정방법을 제안하였다. 이 방법의 타당성을 알아보기 위해 180개의 칼라 도플러 영상에 대해 isotach의 높이와 폭을 추정한 결과, 기존의 에지기반방법이 19개 영상에서 에러를 가지는 반면, 제안한 방법에서는 에러영상이 없음을 알 수 있었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.66-69
• /
• 2017

본 논문에서는 깊은 합성 곱 신경망 (Deep Convolutional Neural Network) 를 이용해서 SAR (Synthetic Aperture Radar) 영상의 반전 잡음 (speckle noise) 성분을 제거하는 기법을 제안하고자 한다. Deep Convolutional Neural Network는 이미지의 데이터 특성에 적합한 딥 러닝 방법이고, 이는 SAR 위성영상의 반전 잡음 제거에 사용해도 효과적이다. 반전 잡음 필터 모델 추정을 위한 학습은 임의로 반전 잡음을 합성한 트레이닝 이미지들과 원본 트레이닝 이미지들을 이용한 회귀모델을 통해 진행된다. 학습을 통해 얻은 반전 잡음 필터는 기존 알고리즘에 비해 우수한 외곽선 보존 성능을 나타냄을 확인하였다.