• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.958-961
• /
• 2019

Convolutional neural network(CNN)의 등장으로 얼굴 표현 인식 연구는 많은 발전을 이루었다. 그러나, 기존의 CNN 접근법은 미리 학습된 훈련모델에서 Multiple-level 의 의미적 맥락을 포함하지 않는 Attention-embedded 문제가 발생한다. 사람의 얼굴 감정은 다양한 근육의 움직임과 결합에 기초하여 관찰되며, CNN 에서 딥 레이어의 산출물로 나온 특징들의 결합은 많은 서브샘플링 단계를 통해서 class 구별와 같은 의미 정보의 손실이 일어나기 때문에 전이 학습을 통한 올바른 훈련 모델 생성이 어렵다는 단점이 있다. 따라서, 본 논문은 Backbone 네트워크의 Multi-level 특성에서 Channel-wise Attention 통합 및 의미 정보를 포함하여 높은 인식 성능을 달성하는 Densely Backwarnd Attention(DBA) CNN 방법을 제안한다. 제안하는 기법은 High-level 기능에서 채널 간 시멘틱 정보를 활용하여 세분화된 시멘틱 정보를 Low-level 버전에서 다시 재조정한다. 그런 다음, 중요한 얼굴 표정의 묘사를 분명하게 포함시키기 위해서 multi-level 데이터를 통합하는 단계를 추가로 실행한다. 실험을 통해, 제안된 접근방법이 정확도 79.37%를 달성 하여 제안 기술이 효율성이 있음을 증명하였다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.17 no.10 s.113
• /
• pp.976-984
• /
• 2006

OFDM system is very useful for the high speed communication system. However, OFDM system has a serious problem of high PAPR that results from the so many subcarriers in the same phase. This OFDM signal is distorted through the nonlinear HPA(High Power Amplifier). Tone reservation method is to insert tone signal in several types to reduce the PAPR after iterating this process by changing the tone signal. Also discrete cosine transform(DCT) can reduces the PAPR as multiplying the cosine value to change the angle and mix up with the data. In the paper, the combination of the TR method and DCT method is newly proposed for more effective reduction of the PAPR. Simulation results show that the proposed method outperforms the conventional simple TR method and DCT method with respect to the PAPR reduction and BER performance.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.4 s.304
• /
• pp.51-58
• /
• 2005

This paper proposes a robust scene change detection technique that uses the weighted chi-square test and the automated threshold-decision algorithms. The weighted chi-square test can subdivide the difference values of individual color channels by calculating the color intensities according to NTSC standard, and it can detect the scene change by joining the weighted color intensities to the predefined chi-square test which emphasize the comparative color difference values. The automated threshold-decision at algorithm uses the difference values of frame-to-frame that was obtained by the weighted chi-square test. At first, The Average of total difference values is calculated and then, another average value is calculated using the previous average value from the difference values, finally the most appropriate mid-average value is searched and considered the threshold value. Experimental results show that the proposed algorithms are effective and outperform the previous approaches.

• Journal of Advanced Navigation Technology
• /
• v.7 no.1
• /
• pp.59-71
• /
• 2003

The channel scheduler is suggested the radio resource management method in order to provide service with guaranteeing fairness and throughput to the users who use limited wireless channel. Proportional fairness scheduling algorithm is the channel scheduler used in the AMC(Adaptive Modulation and Coding)/TDM system, and this algorithm increases the throughput considering the user's time fairness. In this paper is suggested the channel scheduler combining CDM scheme available in AMC/TDM/CDM system. Unlike the system which only uses TDM which provide the only one user at the same slot, this scheduler can service a lot of users since this uses the CDM scheme with multi-cord channel. At every moment, allocation of transmission power to multi-channel users is problematic because of CDM scheme. In this paper, we propose a water-filling scheduling algorithm to solve the problem. Water-filling fairness(WF2) scheduling algorithm watches the average channel environment. So, this modified method guarantees fairness for each user in terms of power and service time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12C
• /
• pp.1044-1051
• /
• 2010

In this paper, we propose an efficient soft-output signal detection method for spatially multiplexed multiple input multiple output (MIMO) systems. The proposed method is based on the ordered successive interference cancellation (OSIC) algorithm, but it significantly improves the performance of the original OSIC algorithm by solving the error propagation problem. The proposed method combines this enhanced OSIC (ESIC) algorithm with a multiple ordering technique in a very efficient way. As a result, the log likelihood ratio (LLR) values can be computed by using a very small set of candidate symbol vectors. The proposed method has been implemented with a $0.13{\mu}m$ CMOS technology for a $4{\times}4$ 16-QAM MIMO system. The simulation and implementation results show that the proposed detector provides a very good solution in terms of performance and hardware complexity.

• Proceedings of the IEEK Conference
• /
• 2001.06d
• /
• pp.45-48
• /
• 2001

Unequal Error Protection(UP) is reasonable scheme in transmission of compressed video with low bit rate. Because it offers the error correction ability each other data according to the Source Significance Information. Hence it can also be flexible to the given channel environment on the video transmission. This paper propose the joint source channel coding through the UEP in consideration of the hierarchical structure of H.263+ based video and the influence of the transmission error. It especially proposes the error-resilient video transmission technique which can reduce complexity of channel coder & decoder by partitioning the video data with a frame. As the result of the proposed algorithm, it is possible to increase the quality of reconstructed video in the error environment without creating additional bits.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1521-1522
• /
• 2011

뇌임펄스에 의한 절연유체 내 절연파괴 현상은 스트리머 채널에서의 유체유동과 기체유동이 동시에 발생한다. 스트리머 개시와 동시에 발생하는 기포들은 절연파괴에 직접적인 영향을 미치며 이를 모의해석하기위해서 2상유동이 고려되어야한다. 2상유동이 고려된 뇌임펄스 응답 유한요소해석은 전계에 의한 푸아송 방정식과 전자, 양이온, 음이온에 대한 전하연속방정식으로 구성되며 전계 방출과 열전자 방출효과를 경계조건으로 부여하였다. 기체 전리현상은 타운젠트 이론을 도입하였으며, 유체 전리현상은 제너 이온화 모델을 도입하여 수학적 모델링을 통한 2상유동으로 결합하였다. IEC standard #60897의 표준규격에 따라 침-구형 전극을 설계하였고 2차원 축대칭 간략화모델에 적용하여 실험적 결과와 비교분석함으로써 신뢰할 수 있는 수치해석기법이 제시되었다.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.706-712
• /
• 2007

국가간 연구목적으로 개설된 초고속 광 네트워크 기술의 발전은 새로운 응용 기술의 등장을 요구하고 있다. 고화질 저지연의 실감 협업 응용은 이러한 연구 목적에 부합할 뿐만 아니라 향후 커뮤니티 기반 응용 기술의 요구를 충족시킬 것으로 보인다. 본 논문에서는 실감 협업 응용 기술에 필요한 비압축 HD stereoscopic 영상 전송 시스템을 구성해 3D HD 영상을 사용자가 체감할 수 있도록 한다. 또한, 소프트웨어 기반 다채널 음성 재생을 다루며 실험을 통해 방향성 있는 협업 환경 구축의 가능성을 보여준다. 입체감 있는 미디어 재생을 위해 병렬 구조의 좌 우 송수신 시스템을 구축 후 stereoscopic 비압축 영상 전송을 수행하며, 좌 우 영상 세션간의 인터 미디어 동기화 기법의 설계방법을 제안한다. 음성 재생 소프트웨어는 ALSA를 이용하여 구현하였으며 가변 데이터 길이 및 프레임 손실로 인한 채널 뒤섞임(channel swapping)을 방지하기 위한 버퍼를 재생 모듈 전처리단에 추가하였다. 초고속 네트워크와 비압축 미디어 전송의 결합은 IP를 이용해 다채널 음성 지원의 실감 HDTV를 가능케 하므로 이를 유용하게 활용할 수 있는 사용 시나리오를 살펴본다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.12
• /
• pp.1401-1412
• /
• 1992

On band limited channel like satellite communication and voice communication. TCM(Trellis Coded Modulation) is a communication method that has coding gain which combines modulation with channel coding without bandwidth expansion. In this paper, we apply PAM, PSK, QAM to TCM, and propose the extended path back method decoding algorithm which improved drawback of viterbi decoding algorithm and apply to TCM this decoding algorithm. Using Monte Carlo simulation, we analyze performance of each modulation technique and efficiency of decoding algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.8A
• /
• pp.592-599
• /
• 2009

In this paper, we consider multiuser multi-input/multi-output antenna systems with zero-forcing receivers in downlink. In this case, to exploit multiuser diversity, spatial-division multiple access (SDMA) system allows to assign different users to a part of transmit antennas at the base station whereas spatial-division multiplexing (SDM) system assigns all antennas to single user's data stream. In this paper, we present analytical frameworks to evaluate performance of these systems. We first analyze the performance of these two systems by deriving closed-form expressions of achievable throughput. Numerical results show that the derived expressions are very tight. In addition, we approximate the capacity expression of SDM and SDMA systems and compare the SDM with the optimal case.