• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.41-49
• /
• 2024

Speech recognition technology is continuously advancing and widely used in various fields. In this study, we aimed to investigate the impact of speech data quality on speech recognition models by dividing the dataset into the entire dataset and the top 70% based on Signal-to-Noise Ratio (SNR). Utilizing Seamless M4T and Google Cloud Speech-to-Text, we examined the text transformation results for each model and evaluated them using the Levenshtein Distance. Experimental results revealed that Seamless M4T scored 13.6 in models using data with high SNR, which is lower than the score of 16.6 for the entire dataset. However, Google Cloud Speech-to-Text scored 8.3 on the entire dataset, indicating lower performance than data with high SNR. This suggests that using data with high SNR during the training of a new speech recognition model can have an impact, and Levenshtein Distance can serve as a metric for evaluating speech recognition models.

• Proceedings of the KOSOMBE Conference
• /
• v.1995 no.05
• /
• pp.42-44
• /
• 1995

The original DANTE sequence and its variations have limitation in excitation profile (a sinc function-like excitation) due to the finite duration of the DANTE pulsetrain. This sinc function-like selection profile excites only a small fraction of the spins in the pixel thereby results in poor signal to noise ratio (only about ${\sim}1%$ of normal MR imaging sequence). Therefore, this poor signal to noise ratio (SNR) has been the main drawback of the original DANTE sequence. To improve the signal to noise ratio, phases of individual RF pulses in the DANTE pulse train were modulated so that more spins in the object were excited ($1{\sim}3$). We have introduced a new FM (Frequency Modulation) DANTE sequence and analyzed the signal intensity and excitation profiles.

• Journal of Satellite, Information and Communications
• /
• v.10 no.4
• /
• pp.46-51
• /
• 2015

In this paper, I design a linear array antenna simulator for anti-jamming GPS systems and perform various performance analysis by simulation. First, I generate simulated transmission signals through the analysis of GPS satellite signal structure, and analyze SNR(Signal to Noise power Ratio) performance of linear array antenna according to number of arrays under noise environments. In addition, I analyze the performance of the anti-jamming beam pattern using MMSE(Minimum Mean Square Error) signal processing method, and also analyze the anti-jamming performance considering antenna calibration problem when there are different delays between arrays.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.5
• /
• pp.507-517
• /
• 2022

The PhotoAcoustic Microscopy (PAM) has been proved to be a useful tool for biological and medical applications due to its high spatial and contrast resolution. PAM is based on transmission of laser pulses and reception of PA signals. Since the strength of PA signals is generally low, not only are high-performance optical and acoustic modules required, but high-performance electronics for imaging are also particularly needed for high-quality PAM imaging. Most PAM systems are implemented with a combination of several pieces of equipment commercially available to receive, amplify, enhance, and digitize PA signals. To this end, PAM systems are inevitably bulky and not optimal because general purpose equipment is used. This paper reports a PA signal receiving system recently developed to attain the capability of improved Signal to Noise Ratio (SNR) and Contrast to Noise Ratio (CNR) of PAM images; the main module of this system is a low noise, wideband signal receiver that consists of two low-noise amplifiers, two variable gain amplifiers, analog filters, an Analog to Digital Converter (ADC), and control logic. From phantom imaging experiments, it was found that the developed system can improve SNR by 6.7 dB and CNR by 3 dB, compared to a combination of several pieces of commercially available equipment.

• Journal of the Korean Society of Radiology
• /
• v.11 no.7
• /
• pp.619-625
• /
• 2017

The sudy was intended to evaluate the optimal equipment selection by quantitatively assessing the SNR(signal to noise ratio) and CNR(contrast to noise ratio) on the abdominal organ. This study performed on 1.5 T and 3.0 T MRI units focusing on HASTE, HASTE(f/s) and FFE(in of phase), FFE(out of phase) without using the contrast medium(Gadolinium). The data analysis was performed by randomly selecting on 1.5 T and 3.0 T abdominal MRI images. As a results, SNR and CNR values of 3.0 T is higher than 1.5 T at liver, kidney and spleen(p<0.05). Stomach, abdominal fat and pancreas was obtained a higher value at 1.5 T(p<0.05). On conclusion, the organs of outer part in the body showed generally a high value at 3.0 T, and the organs of inner part in the body including the gas showed a high value at 3.0 T because of a large difference on magnetic susceptibility.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.11C
• /
• pp.880-882
• /
• 2008

This paper presents LDPC codes' upper bounds over the waterfall SNR region. The previous researches have focused on the average bound or ensemble bound over the whole SNR region and showed the performance differences for the fixed block size. In this paper, the particular LDPC codes' upper bounds for various block sizes are calculated over the waterfall SNR region and are compared with BP decoding performance. For different block sizes the performance degradation of BP decoding is shown.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.14 no.11
• /
• pp.1149-1155
• /
• 2003

OFDM/SFH(orthogonal frequency division multiplexing/slow frequency hopping) system is the combination of OFDM communication system with frequency hopping(FH) method and OFDM/DS(orthogonal frequency division multiplexing/direct sequence) system combines the OFDM communication system with direct sequence(DS) method in terms of PN(pseudo noise) sequence. These two systems are also can be used for anti-jamming. In this study we analyze performances of OFDM/SFH system and OFDM/DS system with the equal information rate in partial band jamming and tone jamming channel. We find BER(bit error rate) in the variation of JFR(jamming fractional ratio) and JSR(jamming to signal power ratio) of the partial band jamming. Also, BER is found in the several JSR of the single tone or multi tone jamming. OFDM/DS system shows better performance than OFDM/SFH system in partial band jamming environment. OFDM/DS system has about 2.5 ㏈ SNR(signal to noise power ratio) gains than OFDM/SFH system when JFR=4/16 and JSR=0 ㏈ to meet BER=10$\^$-3/. However, OFDM/SFH system has about 3 ㏈ SNR gains than OFDM/DS system at single tone jamming of JSR=5 ㏈. In multi tone jamming, performance of OFDM/SFH system is considerably degraded than OFDM/DS system.

• Journal of the Korean Society of Radiology
• /
• v.12 no.7
• /
• pp.845-852
• /
• 2018

This study aimed to investigate the difference of X-ray exposure by comparing and analyzing absorbed dose according to changes in the number of frames in coronary angiography, also depending whether the zoom mode is FOV enlargement or Zoom Live. Moreover, for appropriate frame selection measures for examination, including the effect of frame change on the image quality, were sought by measuring the noise strength expressed by the standard deviation (SD), the signal to noise ratio (SNR) and contrast to noise ratio (CNR). The study was conducted with an anthropomorphic phantom on an angio-system. The linear relationship between the frame rate and the radiation dose was evident. On the contrary, the indices of image quality (SD, SNR, and CNR) were almost constant irrespective of the number of frames. The difference depending on the zoom mode was not statistically significant for DAP, air kerma, and SD (p > 0.05). However, SNR and CNR were statistically different between FOV enlargement and Zoom Live. In conclusion, since the image quality was not degraded significantly with the decreasing frame rate from 30, 15, to 7.5 f/s and the radiation dose evidently decreases in almost exactly linear proportion to the decreasing frame rate, the number of frames per second needs to be maintained as low as reasonably achievable. As for the dependence on the zooming mode, the Live Zoom mode showed statistically significant improvement in the image quality indices of SNR and CNR and it justifies active use of the Live Zoom mode which enables real-time image enlargment without additional radiation dose.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9C
• /
• pp.683-690
• /
• 2008

By using inherent stopping criteria of LDPC codes, the average number of iterations can be substantially reduced at high signal to noise ratio (SNR). However, we encounter a problem when hybrid automatic repeat request (H-ARQ) systems are applied. Frequent failures of decoding at low SNR region imply that the decoder leaches the maximum number of iterations frequently and thus the decoding complexity increases. In this paper, we propose a combination of stopping criteria using the syndrome weight of tentative codeword. By numerical analysis, it is shown that the decoding complexity of given H-ARQ system is reduced by 70-80% with the proposed algorithms.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.8
• /
• pp.1876-1889
• /
• 2013

Packet-based mobile multimedia services for the Internet differ with respect to their resource requirements, performance objectives, and resource usage efficiencies. Nonetheless, each mobile terminal should support a variety of multimedia services, sometimes even simultaneously. This paper proposes a sub-channel allocation scheme based on multi-level priority for supporting mobile multimedia services in an Orthogonal Frequency Division Multiple Access (OFDMA) system. We attempt to optimize the system for satisfying the Quality of Service (QoS) requirements of users and maximize the capacity of the system at the same time. In order to achieve this goal, the proposed scheme considers the Signal-to-Interference-plus-Noise Ratio (SINR) of co-sub-channels in adjacent cells, the Signal-to-Noise Ratio (SNR) grade of each sub-channel in the local cell on a per-user basis, and the characteristics of the individual services before allocating sub-channels. We used a simulation to evaluate our scheme with the performance measure of the outage probabilities, delays, and throughput.