http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
메모리 補償原理를 利用한 二線加入者 線路에서의 反響 除去
黃燦植 경북대학교 공과대학 1983 工大硏究誌 Vol.12 No.-
Full duplex data transmission over two-wire subscriber lines can be achieved by the application of hybrid couplers. This paper describes an algorithm for implementing an echo canceler due to the imperfect isolation between transmitter and receiver at the same end of a connection. A new type of digital echo canceler using memory compensation principle is presented and the quantization effect is also considered.
황찬식 경북대학교 공과대학 1980 工大硏究誌 Vol.9 No.-
Two-Dimensional adaptive equalization is discussed for quadrature amplitude modulation (QAM) signal. Adaptive equalizer is composed of two tapped delay lines (TDL) for inphase and quadra ature signals, and each of TDL is controlled by two sets of tap weights. The tap weights are adaptively adjusted by the iterative procedure based on the gradient following algorithm for minimum mean-square-error. The system is implemented by digital circuitry. To understand the quantization effect on the convergence of the tap weights, the relation between the convergence and step-size in the iteration is investigated. It is shown that8-bit quantization and multiplication is applicable for 2400Hz binary QAM signal.
Fast Minimum Distortion Codevector Search Algorithm for VQ
황찬식,채종길 대한전자공학회 1994 ISPACS:Intelligent Signal Processing and Communica Vol.1 No.1
A critical problem associated with Uie use of vector quantization(VQ) is the computational complexity incurred in the encoding, where the complexity increases exponentially according to codebook size and limits practical application. In this paper, the initialization of fast searching algorithm using prior comparison is proposed to decrease the coding time. This initialization of fast searching algorithm is adopted to reduce re chance of selecting the inappropriate reference codevector which seems to happen in the fast searching algorithm using prior comparison. The proposed algorithm needs only a small fraction of mathematical operations of exhaustive full search VQ and the reduction of the encoding time is shown in the fast searching algorithm with the proposed initialization, compared with the conventional fast searching algorithm.
황찬식,김강욱,윤정오 경북대학교 전자기술연구소 1998 電子技術硏究誌 Vol.19 No.1
H.263 has a hybrid compression technique using inter-picture prediction and transform coding of the remaining signal to utilize the temporal and spatial redundancies respectively. But the conventional basic mode algorithm of H.263 still has the problem in picture quality due to the high compression ratio. So it has been proposed to include four negotiable coding options: Unrestricted Motion Vectors(UMV), Syntax-based Arithmetic Coding(SAC), Advanced Prediction(ADV) and PB-frames(PB). In this paper, we analyze a performance of H.263 negotiable mode for image class. Experimental results show that blocking artifact is greatly reduced by using ADV mode for class B, C and fewer bits are produced by using SAC mode for all classes and the improvement of PSNR is obtained by using UMV mode for class C due to an extension of the motion vector range.
이선 가입자에서의 기본대역 전송을 위한 새로운 디지탈 반향제법방식
황찬식,심영석 대한전자공학회 1984 전자공학회지 Vol.21 No.2
이선 선로를 통한 full-duplex 데이타 전송에 있어서 hybrid-coupler의 불완전한 송·수신에 따른 반향신호는 데이타 검출에 큰 장애가 되고 있다. 본 논문에서는 산술평균에 의한 반향신호 추정법을 제시하고 이를 적응 디지탈 필터에 의한 추정법과 이론적 해석 및 컴퓨터 시뮬레이션을 통해 비교하였다. 양자화 효과를 고려한 이론적인 해석 결과 산술평균 추정을 사용한 반향제거는 hardware 구성이 간단하면서도 수렴이 매우 빠름을 밝혔다. 특히 제시된 방식은 선로 감쇠량에 관계없이 반향신호를 정해진 시간내에 충분히 제거할 수 있다. A new type of digital echo canceler for two-wire digital transmission is presented. The new principle estimates an echo signal by use of the arithmetic means estimate for each transmitted data pattern, which leads to relatively simple hardware. The principle is compared with adaptive digital filter methods through theoretical analysis and computer simulation. The results show that the proposed method has fast convergence property with respect to its hardware simplicity and that the convergence time is independent of echo level. Quantization effects are also analyzed.
고속데이타 전송에 따른 국내 공중 전화선로의 등화에 관한 연구
황찬식,심영석 경북대학교 전자기술연구소 1988 電子技術硏究誌 Vol.9 No.1
An equalization problem has been investigated when CCITT Rec. v. 29 9600 bps modem is used in the national pulic telephone network. It has been shown by computer simulation that a cable or link equalizer in commercial modem can successfuly reduce the burden of adaptive equalizer for highly dispersive channel, and appeared that in national pulic telephone lines the linear distortion were not prevent the successful operation of v. 29 modem.
HVS가중치를 갖는 LOT/CVQ를 이용한 점진적 영상 전송
황찬식 한국통신학회 1993 韓國通信學會論文誌 Vol.18 No.5
LOT(Lapped Transform)와 HVS(Human Visual System) 가중치를 이용하여 분류벡타 양자화기에 근거한 점진적 영상부호화 방법을 제안하였다. DCT변환을 이용한 기존의 블럭변환 영상부호화는 낮은 전송율에서 블럭화 현상이 심하게 나타난다. 본 논문에서는 영상블럭을 LOT로 변환하고 그 블럭들의 구조적 특성에 따라 네 부류로 분류한 다음 적응비트배정으로 재생 영상화질을 개선하기위해 HVS가중치를 갖는 LOT계수특성에 의해 서브벡타들로 나눈다. 이 서브벡타들을 벡타양자화하여 점진적으로 전송한다. 컴퓨터 시뮬레이션을 통한 부호화시험을 하여 점진적 전송에 의한 LOT/CVQ가 효율적인 전송임을 보이고 PIT/DCTVQ를 이용한 결과와 비교하였다. A progressive image transmission (PIT) scheme based on the classified transform vector quantization (CVQ) technique using the lapped orthogonal transform (LOT) and human visual system (HVS) weighting is proposed in this paper. Conventional block transform coding of images using DCT produces in general undesirable block-artifacts at low bit rates. In this paper, image blocks are transformed using the LOT and classified into four classes based on their structural properties and further divided adaptively into subvectors depending on the LOT coefficient statistics with HVS weighting to improve the reconstructed image quality by adaptive bit allocation. The subvectors are vector quantized and transmitted progressively. Coding tests using computer simulations show that the LOT/CVQ based PIT of images is a effective coding scheme. The results are also compared with those obtained using PIT/DCTVQ. The LOT/CVQ based PIT reduces the block-artifacts significantly.