• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.2
• /
• pp.18-26
• /
• 2015

In this paper, we studied the method for analyzing and estimating the parameters that induce I/Q imbalance in the repeater using direct conversion RF. In repeater, amplitude, phase, and filter mismatch are generated in the receiving-end which converts RF signal to baseband signal. And amplitude and phase mismatch are generated in the transmitting-end which converts baseband signal to RF signal. Accordingly, we modeled the parameters that cause I/Q imbalance in the structure of the repeater in order, and proposed a feedback test structure from the transmitting-end to the receiving-end for estimating the corresponding parameters. By comparing the test transmitting signal and received signal, it is possible to estimate the I/Q imbalance parameters which occurred from mixed components of real and imaginary part. And it was confirmed that I/Q imbalance phenomenon has been properly compensated with estimated parameters at the direct conversion RF repeater.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.2
• /
• pp.145-153
• /
• 2019

Adaptive beamforming methods are known to suppress sidelobes and improve detection performance of weak signal by constructing weight vectors depending on the received signal itself. A standard adaptive beamforming like the MVDR (Minimum Variance Distortionless Response) is very sensitive to mismatches between weight vectors and actual signal steering vectors. Also, a large computational complexity for estimating a stable covariance matrix is required when wideband beamforming for a large-scale array is used. In this paper, we exploit the WBRCB (Wideband Robust Capon Beamforming) method for stable and robust wideband adaptive beamforming of a passive large uniform line array sonar. To improve robustness of adaptive beamforming performance in the presence of mismatches, we extract a optimum mismatch parameter. WBRCB with extracted mismatch parameter shows performance improvement in beamforming using synthetic and experimental passive sonar signals.

• Journal of Digital Convergence
• /
• v.13 no.10
• /
• pp.313-318
• /
• 2015

A traditional speech enhancement methods distort the sound spectrum generated according to estimation of the remaining noise, or invalid noise is a problem of lowering the speech recognition performance. In this paper, we propose a speech detection method that convergence the sound energy distribution process and sound energy parameters. The proposed method was used to receive properties reduce the influence of noise to maximize voice energy. In addition, the smaller value from the feature parameters of the speech signal The log energy features of the interval having a more of the log energy value relative to the region having a large energy similar to the log energy feature of the size of the voice signal containing the noise which reducing the mismatch of the training and the recognition environment recognition experiments Results confirmed that the improved recognition performance are checked compared to the conventional method. Car noise environment of Pause Hit Rate is in the 0dB and 5dB lower SNR region showed an accuracy of 97.1% and 97.3% in the high SNR region 10dB and 15dB 98.3%, showed an accuracy of 98.6%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2010.05a
• /
• pp.155-156
• /
• 2010

`Panoramic image' is called generally 'mosaic image', to create a new image by matching the number of relevant images. In this paper, we are entering data through the camera. Therefore, perspective model was used for recognition of camera parameters, minimizing the discrepancies by proposing measure of discrepancy between the frames. In addition, how to create panoramic images is proposed by fixed reference and time-varying reference in order to create panoramic images.

• Progress in Medical Physics
• /
• v.15 no.2
• /
• pp.77-83
• /
• 2004

The commissioning of a model-based treatment planning system requires many parameters to fit the measured depth doses and transverse profiles. For the commissioning of the Pinnacle$^3$ system, through the Monte Carlo (MC) simulation, the necessary parameters, including the photon spectrum, contaminant electrons, off-axis softening and fluency of photons, were observed. Through the simulation the parameters contained valuable information, but the calculated results of the Pinnacle$^3$ using the MC-derived parameters showed discrepancies with those measured for the off-axis softening and the fluency of photons. Even though the MC calculation produces reasonable values for the commissioning, the thorough physical basis of the Pinnacle$^3$'s commissioning process is needed in order to directly use the MC derived parameters.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.3
• /
• pp.613-620
• /
• 2010

Additive noise and channel distortion strongly degrade the performance of speaker verification systems, as it introduces distortion of the features of speech. This distortion causes a mismatch between the training and recognition conditions such that acoustic models trained with clean speech do not model noisy and channel distorted speech accurately. This paper presents a phase-related feature extraction method in order to improve the robustness of the speaker verification systems. The instantaneous frequency is computed from the phase of speech signals and features from the histogram of the instantaneous frequency are obtained. Experimental results show that the proposed technique offers significant improvements over the standard techniques in both clean and adverse testing environments.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.6 no.6
• /
• pp.489-497
• /
• 2001

The proposed dobule phase locked loop and active multiple interphase reactor are used to eliminate the circular current and the voltage ripples caused by the system parameter unbalance of parallel connected UPSs. In this paper, digital controller for the dobule PLL and active interphase reactor is implemented with ADSP21061 as an aspect of functional convenience.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.584-593
• /
• 2018

The highly efficient operation of induction motors has been studied in the past years. Among the many attempts made to obtain highly efficient operation, Maximum Torque Per Amp (MTPA) controls in induction motor drives were proposed. This method enables induction motor drives to operate very efficiently since it achieves the desired torque with the minimal stator current. This is because the alternate qd induction motor model (AQDM) is a highly accurate mathematical model to represent the dynamic characteristics of induction motors. However, it has been shown that the variation of the rotor resistance degrades the performance of the MTPA control significantly, thus leading to its failure to satisfy the maximum torque per amp condition. To take into consideration the mismatch between the actual value of the rotor resistance and its parameter value in the design of the control strategy, an adaptive MTPA control was proposed. In this work, this adaptive MTPA control is investigated in order to achieve the desired torque with the minimum stator current at multiple operating points. The experimental study showed that (i) the desired torque was accurately achieved even though there was a deviation of the order of 5% from the commanded torque value at a torque reference of 25 Nm (tracking performance), and (ii) the minimum stator current for the desired torque (maximum torque per amp condition) was consistently satisfied at multiple operating points, as the rotor temperature increased.

• The Journal of the Korea Contents Association
• /
• v.8 no.7
• /
• pp.1-8
• /
• 2008

In this paper, an algorithm to compensate the inter-view mismatch in the multi-view video according to the different characteristics of cameras is presented. Interview mismatches make it difficult to merge the multi-view images and decrease the coding performance. So, a preprocessing operation to compensate the mismatches is requisite in the multi-view video coding. It is shown that the mismatch in the outputs of multi-view cameras with different electro-optical transfer functions can be approximated with a linear model of a gain and an offset. In addition, a new algorithm for estimating and compensating the inter-view mismatch based on the detection of the overlapped region is presented. Experimental results using various rectified stereo images show that the proposed method compensates inter-view mismatches more accurately compared to the conventional approach.

• Phonetics and Speech Sciences
• /
• v.12 no.3
• /
• pp.65-71
• /
• 2020

Feature normalization is a method to reduce the effect of environmental mismatch between the training and test conditions through the normalization of statistical characteristics of acoustic feature parameters. It demonstrates excellent performance improvement in the traditional Gaussian mixture model-hidden Markov model (GMM-HMM)-based speech recognition system. However, in a deep neural network (DNN)-based speech recognition system, minimizing the effects of environmental mismatch does not necessarily lead to the best performance improvement. In this paper, we attribute the cause of this phenomenon to information loss due to excessive feature normalization. We investigate whether there is a feature normalization method that maximizes the speech recognition performance by properly reducing the impact of environmental mismatch, while preserving useful information for training acoustic models. To this end, we introduce the mean and exponentiated variance normalization (MEVN), which is a compromise between the mean normalization (MN) and the mean and variance normalization (MVN), and compare the performance of DNN-based speech recognition system in noisy and reverberant environments according to the degree of variance normalization. Experimental results reveal that a slight performance improvement is obtained with the MEVN over the MN and the MVN, depending on the degree of variance normalization.