• Speech Sciences
• /
• v.11 no.1
• /
• pp.7-20
• /
• 2004

In this paper, we propose a two-step noise compensation algorithm in feature extraction for achieving robust speech recognition. The proposed method frees us from requiring a priori information on noisy environments and is simple to implement. First, in frequency domain, the Harmonics-based Spectral Subtraction (HSS) is applied so that it reduces the additive background noise and makes the shape of harmonics in speech spectrum more pronounced. We then apply a judiciously weighted variance Feature Vector Normalization (FVN) to compensate for both the channel distortion and additive noise. The weighted variance FVN compensates for the variance mismatch in both the speech and the non-speech regions respectively. Representative performance evaluation using Aurora 2 database shows that the proposed method yields 27.18% relative improvement in accuracy under a multi-noise training task and 57.94% relative improvement under a clean training task.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.3
• /
• pp.443-449
• /
• 2001

This paper is simulation research to improve speech recognition rates under the noisy environment. We examines recognition ratio based on frequency-weighted HMM together with spectral subtraction. As results, frequency-weighted HMM with scaling coefficients is trained as a minimum error classification criterion, and is presents a higher recognition rates in noisy condition than a conventional method. Furthermore, spectral subtraction method gives 11 to 28% improvements for this frequency-weighted HMM in low SNR, and gives recognition rates of 81.7% at 6dB SNR of noisy speech.

• Korean Journal of Optics and Photonics
• /
• v.7 no.4
• /
• pp.428-433
• /
• 1996

We have demonstrated 200 km non-dispersion shifted fiber(NDSF) transmission of 10 Gb/s signal by using mid-span spectral inversion(MSSI) method as a dispersion compensation technique. We have used four-wave mixing process in dispersion shifted fiber(DSF) to generate a spectrum inverted signal. The spectral inversion efficiency of -26.7dB and signal to noise ratio of 23.0 dB have been achieved. The measured sensitivities at $10^{-9}$ bit error rates (BER) were -28.0 dBm in back to back configuration and -27.0 dBm after transmission of 200 km NDSF.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.11b
• /
• pp.320-323
• /
• 2001

A computer simulation was performed to distinguish quantitatively a color reaction in a urine analysis systems by using the spectral power distribution of LEDs. the spectral reflectance of a urine strip. and the spectral sensitivity of photodiode. The CIE tristimulus values and CIE chromaticity coordinates ware modified to be conformable with real color reactions in a urine strips. Results on color simulation showed a of real color in comparison with those obtained by Colorimeter CM2C(Color Savvy).

• Journal of Advanced Navigation Technology
• /
• v.10 no.2
• /
• pp.159-167
• /
• 2006

In this paper the numerical methods of finding the optimal position of optical phase conjugator (OPC) and the optimal fiber dispersion are proposed, which are able to effectively compensate overall channels in $16{\times}40$ Gbps WDM system. And the compensation characteristics in the system with two induced optimal parameters are compared with those in the system with the currently used mid-span spectral inversion (MSSI) in order to confirm the availability of the proposed methods. It is confirmed that the reception performances are largely improved in the system with the induced optimal parameters than in the system with MSSI through the analyzing the eye opening penalty (EOP) and bit error rate (BER) characteristics. It is also confirmed that two optimal parameters depend on each other, but are less related with the procedural problem about the first optimal value among these.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.11A
• /
• pp.1044-1052
• /
• 2006

In this paper the optimum position of optical phase conjugator (OPC) and the optimal dispersion coefficients of filler sections in WDM system with the conventional mid-span spectral inversion (MSSI) are numerically induced and then applied into $16{\times}40$ Gbps WDM systems with 1,000km non zero - dispersion shifted fiber(NZ-DSF) in order to efficiently compensate for the distorted overall channels. It is confirmed that the compensation extents of overall WDM channels are more improved by applying the induced optimal parameters into WDM system than those in WDM system with the conventional MSSI. So it is expected to alternate with the forming method of the symmetrical distributions of power and local dispersion by applying these optimal parameters into the real WDM system, which generate a serious problem of applying the OPC into multi-channels WDM system if it is not formed. It is also confirmed that two optimal parameters depend on each other, but less related with the finding procedure. And, it will be possible to realize the flexible system design by applying the methods proposed in this paper into the real WDM system with OPC.

• Journal of information and communication convergence engineering
• /
• v.5 no.1
• /
• pp.17-22
• /
• 2007

In this paper, the optimal value of optical phase conjugator (OPC) position and the optimal values of dispersion coefficients of fiber sections for the best compensation of the distorted WDM signals with frequency chirp of -1 are induced to alternate with the symmetrical distributions of power and local dispersion with respect to OPC, which is difficult to form in real optical link due to fiber attenuation in mid-span spectral inversion (MSSI) technique. It is confirmed that the Q-factors of total channels of -18.5 dBm launching light power exceed 16.9 dB, which value corresponds to 10-12 BER, by applying the induced optimal parameter values into 16 channels ${\times}$ 40 Gbps WDM system, on the other hand the Q-factors of only 9 channels exceed that value in WDM system with the conventional MSSI technique. Thus, it is expected to expand the availability of OPC in WDM system through the using of the optimal parameter values that are induced by the proposed method in this paper, without the symmetrical distributions of power and local dispersion.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.13 no.7
• /
• pp.716-725
• /
• 2002

In this paper, we showed the applicability of power symmetric MSSI(Mid-Span Spectral Inversion) to the long-haul WDM system with the channel interference due to FWM(Four Wave Mixing). And we showed the degree of performance improvement. We used 1 dB EOP(Eye-Opening Penalty) criterion so as to evaluate the degree of compensation dependent on the variation of chirp parameter of optical pulse for the various input power in high speed tansmission system. And we evaluated the maximum input power of channel be able to be the signal to crosstalk noise (SNR) above 20 dB in the transmission link with the channel interference due to FWM. Consequently the proposed MSSI compensation method is capable to transmitting the total 68 WDM channels simultaneously with a 0.4 nm channel spacing and 5.3 dBm maximum input power in a 10 Gbps transmission link. Therefore the proposed power symmetric MSSI compensation method may be very useful for the implementation of long-haul wideband WDM transmission systems with relatively high power and improved performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.12A
• /
• pp.1225-1232
• /
• 2008

In this paper, we describe a compensation method of diurnal effect which is one of the factors giving large effect on the performance when using ground-wave signals like Loran-C for a backup and substitute navigation system of global satellite navigation system such as GPS, and currently many researches of the topics are doing in USA and in Europe. In order to compensate diurnal effect, we find periodic frequency components by using the Least Square Spectral Analysis (LSSA) method at first and then compensate the effect by subtracting the estimated compensation signal, obtained by using the estimated amplitude and phase of the individual frequency component, from the original signal. In this paper, we propose a simple compensation algorithm and analysis the performance through simulations. From the results, it is observed that the amplitude and phase can be estimated with under 5 % and 0.17 % in a somewhat poor receiving situation with 0 dB Signal to Noise Ratio (SNR). Also, we analyze the obtainable performance improvement after compensation by using the measured Loran-C data. From the results, it is observed that we can get about 22 % performance improvement when a moving average with 5 minutes interval is employed.

• Journal of the Optical Society of Korea
• /
• v.15 no.4
• /
• pp.386-393
• /
• 2011

In this study, we report an effective k-domain linearization method with a pre-calibrated indexed look-up table. The method minimizes k-domain nonlinear characteristics of a swept source optical coherence tomography (SS-OCT) system by using two arrays, a sample position shift index and an intensity compensation array. Two arrays are generated from an interference pattern acquired by connecting a Fabry-Perot interferometer (FPI) and an optical spectrum analyzer (OSA) to the system. At real time imaging, the sample position is modified by location movement and intensity compensation with two arrays for linearity of wavenumber. As a result of evaluating point spread functions (PSFs), the signal to noise ratio (SNR) is increased by 9.7 dB. When applied to infrared (IR) sensing card imaging, the SNR is increased by 1.29 dB and the contrast noise ratio (CNR) value is increased by 1.44. The time required for the linearization and intensity compensation is 30 ms for a multi thread method using a central processing unit (CPU) compared to 0.8 ms for compute unified device architecture (CUDA) processing using a graphics processing unit (GPU). We verified that our linearization method is appropriate for applying real time imaging of SS-OCT.