http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
QMF를 이용한 대역 압축 시스템에서의 스크램블링 효과
주언경,蔡鍾吉 慶北大學校 1992 論文集 Vol.52 No.-
Scrambling has been used as a method to encrypt the transmission signal for purpose of service protection from illegal use of unauthorized subscribers in Conditional Access system. More recently it is proposed to be used in a HDTV system such as MIT-CC in order to turn ghost, correlated noise, to uncorrelated white noise. In this paper, band rotation scrambling method using a modified line rotation is proposed for the band-compressed system using QMF. And scrambling effects on conversion of ghost to white noise and service protection are studied and analyzed. As results, effects of whitening ghost and service protection by the proposed method are shown to be superior to the existing line rotation method.
주언경,김정구 경북대학교 전자기술연구소 1993 電子技術硏究誌 Vol.14 No.2
In this paper, detection performances are investigated and compared among the combinations of pulse signals transmitted, tares models and target detectors. Incoherent detector using LFM pulse signal is shown to be superior to other detectors in terms of detection performance and hardware complexity for Swelling 0, 1 or 3 targets. For Swerling 2 or 4 targets, semicoherent detector using SFM pulse signal performs better than other detectors if a single pulse is used. On the other hand when multiple pulses are used, incoherent detector using LFM pulse signal has more advantage in detection performance and hardware complexity than coherent and semicoherent detectors.
Hamming (7, 4) 부호를 이용한 오류정정선로 부호의 복호비트오율 분석
주언경,김정구,김영수 경북대학교 전자기술연구소 1994 電子技術硏究誌 Vol.15 No.1
In this paper, residual bit error rates of error correcting codes(ECLC) based on Hamming (7,4) code are calculated analytically, and more accurate approximation coefficients for the rate are suggested. Since encoding and decoding of ECLC is performed with the same process of error correcting code (ECC), residual bit error rate of ECC is obtained. And with the results of ECC, residual bit error rate of ECLC is obtained by considering the errors in the information and line bits. In general, the results show higher residual bit error rate in ECLC than ECC because of the error effect in line bit. And between the ECLC's based on the same ECC, the one with the lower code rate shown lower residual bit errer rate.
Design of error-correcting balanced codes based on partition chain
주언경,정창기 에스케이텔레콤 (주) 1999 Telecommunications Review Vol.9 No.5
In this paper, error-correcting balanced codes based on partition chain are presented. The partition chain consisting of balanced vectors only can be constructed from the code partition chain of Reed-Muller codes. For the partition chain, its line coding parameters such as the maximum runlength and running digital sum are obtained. The trellis and multilevel code structure can be used to design the error-correcting balanced codes. As results, the errorcorrecting balanced codes based on the partition chain can be designed simply and they have better coding parameters than the conventional codes.
주언경,채종길 경북대학교 전자기술연구소 1992 電子技術硏究誌 Vol.13 No.1
In DCT image coding, the quantization step-sizes are varied according to transform coefficients. And they ark reconstructed by multiplying the received quantization level and standard deviation of each coefficient which is received from the coder or estimated in the decoder. However, since quantization step-sizes are inversely proportional to the standard deviation of transform coefficient, a single step-size uniform quantizer can be used in the coder. Then, the hardware can be simplified because reception or estimation of standard deviations used for normalization in the course of quantiation is not required. In this paper, DCT image coding method using a single step-size uniform quantizer mentioned above is proposed and its performances are analyzed and compared with those of the existing method.
주언경,蔡鍾吉 慶北大學校 1992 論文集 Vol.52 No.-
DCT coding, using energy compaction and decorrelation by transform, is proved to be efficient in the absence of channel error, but its performance degrades rapidly over noisy channel. In this paper, bit selective error correction coding technique is presented for protecting DCT coefficients over noisy channel. Since each bit transmitted in a block has different error sensitivity, bit error sensitivity is computed and analyzed for various binary codes, images and DCT block sizes. Then the bits with higher error sensitivity are selected to be protected by BCH error correction codes. As a result, without increasing transmission bits, PSNR can be improved up to 8-10dB and the image with improved quality is reconstructed.
On Initial Value Estimation of the Lloyd-Max Algorithm
Joo, Eon-Kyeong 경북대학교 산업기술연구소 1988 産業技術硏究誌 Vol.16 No.-
아날로그 신호를 디지탈 신호로 변환시키는 과정에서는 필수적으로 양자화가 필요하다. 따라서 이를 최적화 하는 방식에 대해 많은 연구가 진행되어 왔으며 그 중 가장 간단한 방식으로 최소평균자승오차 양자화방식이 있다. 그러나 이 방식도 그 알고리즘(Lloyd-Max Algorithm)이 반복계산을 이용함으로서 그 초기치의 정확도에 의해 최적 설계치수를 구하는 수렴도가 크게 좌우된다. 본 논문에서는 현재까지 발표된 방식들을 검토하여 새롭고 효율적인 초기치 추정방식을 제안하고 그 성능을 검토하였다.
배상재,주언경,Bae, Sang-Jae,Ju, Eon-Gyeong 대한전자공학회 2002 電子工學會論文誌-TC (Telecommunications) Vol.39 No.3
In this paper, a new concatenation scheme of serial concatenated convolutional codes is proposed and the performance analyzed. In the proposed scheme, each of information and parity bits of outer code is entered into inner code through interleaver and deinterleaver. Therefore, the interleaver size is same as the length of input frame. Since the interleaver size of proposed type is reduced to half of the conventional Benedetto type, the interleaver delay time required for iterative decoding is reduced. In addition the multiplexer and demultiplexer are not used in the decoder of the proposed type, the complexity of decoder can be also reduced. As results of simulation, the performance of proposed type shows the better error performance as compared to that of the conventional Benedetto type in case of the same interleaver size. And it can be observed that the difference of BER performance is increased with the increase of Eb/No. In case of the same length of input frame, the proposed type shows almost same performance with Benedetto type despite that the interleaver size is reduced by half. 본 논문에서는 직렬연접 길쌈부호의 새로운 연접방법을 제시하고 그 성능을 분석하였다. 제시된 직렬연접 길쌈부호에서는 외부부호의 정보비트와 패러티비트가 각각 인터리버와 역인터리버를 거쳐서 내부부호로 입력된다. 따라서 인터리버와 역인터리버의 크기는 입력프레임의 크기와 동일하다. 제안된 직렬연접 길쌈부호에서는 인터리버의 크기가 Benedetto 구조에 비하여 반으로 줄어들기 때문에 반복복호시 매번 거쳐야 하는 인터리버의 지연시간을 줄일 수 있다. 또한 제시된 구조에서는 다중화기와 역다중화기가 사용되지 않기 때문에 복호기의 복잡도가 Benedetto 구조보다 상대적으로 더 간단하다. 모의실험 결과 인터리버의 크기가 동일한 경우 제시된 직렬연접 길쌈부호의 성능이 Benedetto 구조의 성능보다 더 우수하였다. 또한 Eb/No가 증가할수록 제시된 구조와 Benedetto 구조의 성능차이가 더욱 증가하였다. 입력프레임의 크기가 동일한 경우에는 제시된 직렬연접 길쌈부호의 성능이 인터리버의 크기가 반으로 줄어듦에도 불구하고 Benedetto 구조와 거의 동일한 성능을 나타내었다.