• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.33-36
• /
• 2007

In this paper, we investigate a fast speaker adaptation method based on eigenvoice in several noisy environments. In order to overcome its weakness against noise, we propose a noisy environment clustering method which divides the noisy adaptation utterances into utterance groups with similar environments by the vector quantization based clustering using a cepstral mean as a feature vector. Then each utterance group is used for adaptation to make an environment dependent model. According to our experiment, we obtained 19-37 % relative improvement in error rate compared with the simultaneous speaker adaptation and environmental compensation method

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.12 s.354
• /
• pp.61-67
• /
• 2006

This paper proposes the algorithm for tracking of the residual phase errors incurred by carrier frequency offset and sampling frequency offset in the orthogonal frequency division multiplexing (OFDM) systems which are suitable for high data rate wireless communications. In the OFDM systems the subcarriers which are orthogonal to each other are modulated by digital data and transmitted simultaneously. The carrier frequency offset causes degradation of signal to noise ratio(SNR) performance and interference between the adjacent subcarriers. The errors in the sampling timing caused by the sampling frequency difference between the transmitter and the receiver sides also cause a major performance degradation in the OFDM systems. The residual error tracking and compensation mechanism is essential in the OFDM system since the carrier and the sampling frequency offset cause the loss of orthogonality resulting in the system performance loss. This paper proposes the scheme where the channel gain and the payload data information are reflected in the residual error tracking process which results in the reduction of the estimation error and the tracking performance improvements under the frequency selective fading wireless channels.

• Phonetics and Speech Sciences
• /
• v.4 no.3
• /
• pp.95-101
• /
• 2012

In this paper, we introduce the automatic speaker identification system 'SPO(Supreme Prosecutors Office) Verifier'. SPO Verifier is a GMM(Gaussian mixture model)-UBM(universal background model) based automatic speaker recognition system and has been developed using Korean speakers' utterances. This system uses a channel compensation algorithm to compensate recording device characteristics. The system can give the users the ability to manage reference models with utterances from various environments to get more accurate recognition results. To evaluate the performance of SPO Verifier on Korean speakers, we compared this system with one of the most widely used commercial systems in the forensic field. The results showed that SPO Verifier shows lower EER(equal error rate) than that of the commercial system.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1343-1345
• /
• 2022

본 논문에서는 다해상도 피라미드 피처 맵 압축 손상 보상 방법을 제안한다. 본 논문에서 제안하는 방법은 패킹된 C-레이어 피처 맵을 비디오 코덱으로 압축할 때, 저해상도 계층의 원본 피처 맵과 복원된 피처 맵 간의 차분 값을 구해 이를 고해상도 계층의 피처 맵에 더해줌으로써 부호화 과정에서 발생하는 오차를 보상하는 방법이다. 본 논문에서 제안하는 방법의 성능을 평가하기 위하여 OpenImageV6 데이터셋 중 1000 장에 대해 객체 검출 성능을 평가하였다. 본 논문에서 제안하는 피처 맵 압축 방법은 C-레이어 피처 맵 압축 방법 대비 bpp 와 mAP 의 BD-rate 관점에서 35.10%의 성능 향상을 보인다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.14 no.8
• /
• pp.845-854
• /
• 2003

In this paper, the characteristics of compensation for WDM channel signal distortion due to both chromatic dispersion and Ken effect in 1,000 km 200 Gbps(5${\times}$40 Gbps) WDM systems was investigated. The WDM system has a path-averaged intensity approximation(PAIA) mid-span spectral inversion(MSSI) as a compensation method. This system has a highly nonlinear dispersion shifted fiber(HNL-DSF) optical phase conjugator(OPC) in the mid-way of transmission line. In order to evaluate the degree of compensation, 1 dB eye opening penalty(EOP), bit error rate(BER) characteristics and power penalty of 10$\^$-9/ BER are used. It is confirmed that HNL-DSF is an useful nonlinear medium in OPC fur wideband WDM system with PAIA MSSI and that the optimal compensation for WDM channel distortion is achieved by the selection of pump light power of OPC, which equalize the conjugated light power into the second half fiber section with the input WDM signal light power depending on total transmission length, dispersion coefficient of fiber, OPC pump light wavelength, conversion efficiency of WDM channel in OPC.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.2
• /
• pp.295-304
• /
• 1996

Toprovide economical BISDN service, with which integration process of many types of media is possible, it is necessary to construct a system with ground network and satellite network combined. The method for this type of transmission using satellite is TDMA that can provide services to many users in various area. However, the most difficult task to connect TDMA which uses synchronous method to ATM which used asynchronous transfer mode is the deterioration n of ATM transmission quality such as cell delay variation. Therefore, it is necessary to develop delay variation compensation method which can confront to the ATM. Efficient ways to use satellite links under the conditions such that maximum efficiency of the delay variation is limited under the required value, and the burst characteristic of transmission cell does not increase are being researched for translation between in ATM and TDMA. This paper points out the problems when time stamp method, reviewd in ground network, is applied to the satellite links to compensate the delay variation .To solve the problem, discrete cell count method is introduced along with the calculation of transmission capacity and error rate.Also, from the observation of stab-ility of the system and verification of reliability even when singal error occurred in the cell transmission timing information, the proposed compensation method appeared to be excellent.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.120-121
• /
• 2006

In this paper, We have designed a face recognition system based on the embedded Linux. This paper has an aim in embedded system to recognize the face more exactly. At first, the contrast of the face image is adjusted with lightening compensation method, the skin and lip color is founded based on YCbCr values from the compensated image. To take advantage of the method based on feature and appearance, these methods are applied to the eyes which has the most highly recognition rate of all the part of the human face. For eyes detecting, which is the most important component of the face recognition, we calculate the horizontal gradient of the face image and the maximum value. This part of the face is resized for fitting the eye image. The image, which is resized for fit to the eye image stored to be compared, is extracted to be the feature vectors using the continuous wavelet transform and these vectors are decided to be whether the same person or not with PNN, to miminize the error rate, the accuracy is analyzed due to the rotation or movement of the face. Also last part of this paper we represent many cases to prove the algorithm contains the feature vector extraction and accuracy of the comparison method.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2S
• /
• pp.701-710
• /
• 2000

This paper presents a new method for an efficient coding of very low bit-rate color video based on adaptive wavelet transform. Our approach reveals that the coding process works more efficiently if the quantized wavelet coefficients are preprocessed by a mechanism exploiting the redundancies in the wavelet subband structure. Thus, we focuses optimized activity of coding part, and exhaustive overlapped block motion compensation is utilized to ensure coherency in motion compensated error frames, and raised cosine window is applied. The horizontal and vertical components of motion vectors are encoded separately using adaptive arithmetic coding while significant wavelet coefficients are encoded in bit-plane order using adaptive arithmetic coding. On average the proposed codec exceeds H.263 and ZTE in peak signal-to-noise ratio by as much as 2.07 and 1.38dB at 28 kbits, respectively. Fore entire sequence coding, 3DWCVC method is superior to H.263 and ZTE by 0.35 and 0.71dB on average, respectively.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.28 no.12
• /
• pp.948-953
• /
• 2017

When using a high-power amplifier(HPA) for high-speed communication, the nonlinear characteristics of the HPA deteriorate power efficiency, bit error rate(BER) performance, and spectral efficiency. Because it is inevitable to use the HPA to obtain sufficient transmission power for high-speed communication, it is necessary to compensate for nonlinearity of the HPA by using a predistorter. In this study, a predistorter was used to compensate for the nonlinearity of the HPA, and the nonlinear distortion was compensated using the predistorter. Simulation results show that the compensation of the nonlinearity of the HPA using the predistorter achieves a BER performance similar to that of an ideal linear amplifier, and that the spectral mask is also satisfied.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.3
• /
• pp.169-180
• /
• 2023

The performance of underwater acoustic communication is greatly affected by multipath propagation and Doppler spread. This paper proposes a new communication technique, the Triangle Spread Carrier (TSC) technique, by modifying the existing Sweep Spread Carrier (SSC) technique that is strong in a multipath propagation environment. The proposed TSC technique is a form in which the up-chirp and down-chirp signals have repeated carriers, and each correlation function characteristic is used to estimate and correct the Doppler shift frequency of the receiving signal. To demonstrate the performance of the proposed TSC technique, we present the results of simulations using underwater channel simulators and sea trial conducted in the East Sea. When demodulating using only the estimated Doppler shift frequency as a result of the sea trial, the uncoded bit error rate was up to 0.194, but when the proposed method was applied, the uncoded bit error rate was reduced to 0.001.