• JSTS:Journal of Semiconductor Technology and Science
• /
• v.13 no.3
• /
• pp.213-223
• /
• 2013

A unified channel thermal noise model valid in all operation regions is presented for short channel MOS transistors. It is based on smooth interpolation between weak and strong inversion models and consistent physical model including velocity saturation, channel length modulation, and carrier heating. From testing for noise benchmark and comparing with published noise data, it is shown that the proposed noise model could be useful in simulating the MOSFET channel thermal noise in all operation regions.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.10 no.3
• /
• pp.225-231
• /
• 2010

In this paper, we investigate the channel thermal noise in nanoscale MOSFETs. Simple analytical model of thermal noise factor in nanoscale MOSFETs is presented and it is verified with accurately measured noise data. The noise factor is expressed in terms of the channel conductance and the electric field in the gradual channel region. The proposed noise model can predict the channel thermal noise behavior in all operating bias regions from the long-channel to nanoscale MOSFETs. From the measurement results, we observed that the thermal noise model for the long-channel MOSFETs does not always underestimate the short-channel thermal noise.

• Speech Sciences
• /
• v.11 no.1
• /
• pp.43-52
• /
• 2004

This paper concerns an effective dual-channel noise reduction method to increase the performance of speech recognition in a car environment. While various single channel methods have already been developed and dual-channel methods have been studied somewhat, their effectiveness in real environments, such as in cars, has not yet been formally proven in terms of achieving acceptable performance level. Our aim is to remedy the low performance of the single and dual-channel noise reduction methods. This paper proposes an effective dual-channel noise reduction method based on a high-pass filter and front-end processing of the eigendecomposition method. We experimented with a real multi-channel car database and compared the results with respect to the microphones arrangements. From the analysis and results, we show that the enhanced eigendecomposition method combined with high-pass filter indeed significantly improve the speech recognition performance under a dual-channel environment.

• Journal of Communications and Networks
• /
• v.14 no.3
• /
• pp.237-242
• /
• 2012

One of the major problems for accurate channel estimation in power line communication systems is impulsive noise. Traditional channel estimation algorithms are based on the assumption of Gaussian noise, or the need to locate the positions of impulsive noise. The algorithms may lose optimality when impulsive noise exists in the channel, or if the location estimation of impulsive noise is inaccurate. In the present paper, an effective channel estimation algorithm based on a robust cost function is proposed to mitigate impulsive noise. The proposed method can provide a closed-form solution, and the application of robust estimation theory enables the proposed method to be free from localization of impulsive noise and thus can guarantee that the proposed method has better performance. Simulations verified the proposed algorithm.

• Journal of Broadcast Engineering
• /
• v.16 no.2
• /
• pp.207-215
• /
• 2011

A signal passed through multi-path channel suffers ISI(Inter-Symbol Interference) and severe distortions caused by channel delay spread and noise components at the SC-FDE(Single Carrier with Frequency Domain Equalizer) transmission. Conventional UW(Unique-Word) based SC-FDE iterative channel estimation improves channel estimation performance by smoothing estimated CIR(Channel Impulse Response) of the noise components outside the channel length at time domain and restoring the broken cyclic property through UW reconstruction. In this paper, we propose channel estimation scheme through noise suppression within channel length. To suppress the noise, we estimate noise standard deviation as estimated CIR of the noise components outside the channel length and make criteria of the noise standard deviation gain that doesn't affect the original signal samples. When estimated CIR samples within channel length are less than the criteria value using the noise standard deviation and gain, the noise components are removed. Simulation results show that the proposed channel estimation scheme brings good channel MSE(Mean Square Error) and good BER(Bit Error Rate) performance.

• Journal of KSNVE
• /
• v.9 no.6
• /
• pp.1137-1144
• /
• 1999

Most of ANC(active noise control) researches are focused on adaptive algorithms, computer simulations and implementations of single-channel system in experimental environments. In this paper, a multi-channel ANC system based on DSP's was developed to obtain global attenuations over wide region and applied to the active control of the humming noise generated by a transformer. The developed ANC system including 24 microphones and 12 spearkers was applied to the real transformer noise reduction problem. Results showed that the control system could successfully control the humming noise over the region of interest.

• Journal of Broadcast Engineering
• /
• v.13 no.5
• /
• pp.565-573
• /
• 2008

This paper proposes noise reduction methods to improve the channel estimation performance in Equalization On-Channel Repeater (E-OCR) for Terrestrial-Digital Multimedia Broadcasting (T-DMB) system. In the E-OCR for T-DMB system, the channel estimation is performed by the pilot signal extraction and the channel coefficient estimation. Since the performance of the channel estimation highly depends on the estimated channel coefficients whose accuracy is associated with the received signal-to-noise ratio (SNR), it is important to reduce noise existing at the estimated channel coefficients. To reduce such noise components, various noise reduction methods of the estimated channel coefficients based on the M -point weighted moving average are proposed. Computer simulations show that the proposed methods improve the estimation performance by more than 2 to 3 dB in terms of the symbol error rate after equalization.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.1
• /
• pp.154-160
• /
• 2004

In this paper, implementation of active noise curtains using multiple channel adaptive filters is presented. The same numbers of single channel LMS algorithms as control loudspeakers is used instead of a multi-channel LMS algorithm to reduce the computational burden of adaptive filter algorithms. In general, a multi-channel LMS algorithm is usually used in active noise control system. but this algorithm has much more computational complexity. The single channel control techniques have less amount of DSP calculation, compared to multiple channel control techniques. A stabilizing procedure for adaptive IIR filters is also proposed to improve the stability of recursive LMS algorithms. Both experimental results of two control techniques using TMS320VC33 digital signal processor show the similar noise reduction, but the single channel control techniques are more efficient in practical active noise curtain applications

• Transactions of the Korean Society of Automotive Engineers
• /
• v.3 no.6
• /
• pp.23-29
• /
• 1995

In this study, Active noise controlis implemented with single channel and multi-channel adaptive algorithm in 3 dimensional reverberant enclosure sound. field, which occurrs in complicated acoustic mode. First, for the one case excited with the resonant frequency of an enclosure, a target of control and the other cases excited with band-pass filtered random noise(100~400Hz), it is implemented to control reducing interior noise of enclosure with single channel and realtime multi-channel adaptive algorithm for global noise reduction in enclosure.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.5 no.3
• /
• pp.187-194
• /
• 2005

An analysis of the 1/f or flicker noise in FD n-channel Double Gate SOI MOSFET is proposed. In this paper, the variation of power spectral density (PSD) of the equivalent noise voltage and noise current with respect to frequency, channel length and gate-to-source voltage at various temperatures and exponent $C(i.e\;1/f^c$ is reported. The temperature is varied 125 K from to room temperature. The variation of PSD with respect to channel length down to $0.1{\mu}m$ technology is considered. It is analyzed that l/f noise in FD n-channel Double Gate SOI MOSFET is due to both carrierdensity fluctuations and mobility-fluctuations. But controversy still exits to its origin.