• ETRI Journal
• /
• v.45 no.1
• /
• pp.150-162
• /
• 2023

The polar delta-sigma modulator (DSM) transmitter architecture exhibits good coding efficiency and can be used for software-defined radio applications. However, the necessity of high clock speed is one of the major drawbacks of using this transmitter architecture. This study proposes a low-complexity timeinterleaved architecture for the polar DSM transmitter baseband part to reduce the clock speed requirement of the polar DSM transmitter using an upsampling technique. Simulations show that using the proposed four-branch timeinterleaved polar DSM transmitter baseband part, the clock speed requirement of the transmitter is reduced by four times without degrading the signal-tonoise-and-distortion ratio.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.20 no.12
• /
• pp.85-91
• /
• 2021

In our daily life, artificial intelligence performs simple and complicated tasks like us, including operating mobile phones and working at homes and workplaces. Artificial intelligence is used in industrial technology for diagnosing various types of equipment using the machine learning technology. This study presents a fault mode effect analysis (FMEA) of start motors using machine learning and big data. Through multiple data collection, we observed that the primary failure of the start motor was caused by the melting of the magnetic switch inside the start motor causing it to fail. Long-short-term memory (LSTM) was used to diagnose the condition of the magnetic locations, and synthetic data were generated using the synthetic minority oversampling technique (SMOTE). This technique has the advantage of increasing the data accuracy. LSTM can also predict a start motor failure.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.331-337
• /
• 2023

In this paper, we propose a deep learning model to detect lumbar 3 (L3) CT images to determine the occurrence and degree of sarcopenia. In addition, we would like to propose an optimization technique that uses oversampling ratio and class weight as design parameters to address the problem of performance degradation due to data imbalance between L3 level and non-L3 level portions of CT data. In order to train and test the model, a total of 150 whole-body CT images of 104 prostate cancer patients and 46 bladder cancer patients who visited Gangneung Asan Medical Center were used. The deep learning model used ResNet50, and the design parameters of the optimization technique were selected as six types of model hyperparameters, data augmentation ratio, and class weight. It was confirmed that the proposed optimization-based L3 level extraction model reduced the median L3 error by about 1.0 slices compared to the control model (a model that optimized only 5 types of hyperparameters). Through the results of this study, accurate L3 slice detection was possible, and additionally, we were able to present the possibility of effectively solving the data imbalance problem through oversampling through data augmentation and class weight adjustment.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.12 no.3
• /
• pp.278-285
• /
• 2012

A digital readout IC for capacitive sensors is presented. Digital capacitance readout circuits suffer from static capacitance of sensors, especially single-ended sensors, and require large passive elements to cancel such DC offset signal. For this reason, to maximize a dynamic range with a small die area, the proposed circuit features digital filters having a coarse and fine compensation steps. Moreover, by employing switched-capacitor circuit for the front-end, correlated double sampling (CDS) technique can be adopted to minimize low-frequency device noise. The proposed circuit targeted 8-kHz signal bandwidth and oversampling ratio (OSR) of 64, thus a $3^{rd}$-order ${\Delta}{\Sigma}$ modulator operating at 1 MH was used for pulse-density-modulated (PDM) output. The proposed IC was designed in a 0.18-${\mu}m$ CMOS mixed-mode process, and occupied $0.86{\times}1.33mm^2$. The measurement results shows suppressed DC power under about -30 dBFS with minimized device flicker noise.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.4
• /
• pp.1-8
• /
• 2010

In this paper, we propose a new gain optimization technique for Sigma-Delta A/D converters. In the proposed scheme, multiple gain set candidates showing maximum SNR in the modulator block are selected, and then multiple gain set candidates are investigated for minimum MSE in decimation block. Through CIC decimation filter simulation, it is shown that second SNR ranking candidate in modulation block is the best gain set.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.4
• /
• pp.790-796
• /
• 2014

Identification of communication channels is a problem of important current theoretical and practical concerns. Recently proposed solutions for this problem exploit the diversity induced by antenna array or time oversampling. The method resorts to an adaptive filter with a linear constraint. In this paper, an approach is proposed that is based on decomposition. Indeed, the eigenvector corresponding to the minimum eigenvalue of the covariance matrix of the received signals contains the channel impulse response. And we present an adaptive algorithm to solve this problem. Proposed technique shows the better performance than one of existing algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.10A
• /
• pp.1473-1479
• /
• 2000

This paper presents the performance evaluation of an adaptive QAM scheme under flat and frequency selective fading channels for indoor wireless communication systems. The QAM modulation is combined with differential encoding and the demodulation process is carried out noncoherently. The adaptation is performed by varying the modulation level of QAM, depending upon received signal strength. The adaptation mechanism allows a 2- or 3-bit increase or decrease at a time, if the channel condition is considered to be significantly good or bad. Simulation results show that the average number of bits per symbol (ABPS) for each symbol block transmitted over a flat fading channel is higher than 5.0 and the BER performance is better than 10^-4 for a SNR value higher than 30 dB. For frequency selective fading channels, an oversampling technique in the receiver was employed. The BER performance obtained for frequency selective fading channels is better than 10^-4 with a SNR value of 40 dB and ABPS is found to be approximately 5.5. Therefore, this scheme is very useful in that it provides both very high bandwidth efficiency and acceptable performance with moderate SNR values over flat and frequency selective fading channels. In addition, this scheme provides reduced receiver complexity by way of noncoherent detection.

• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.588-600
• /
• 2019

Each year Malaria affects over 200 million people worldwide. Particularly, African continent is highly hit by this disease. According to many researches, this continent is ideal for Anopheles mosquitoes which transmit Malaria parasites to thrive. Rainfall volume is one of the major factor favoring the development of these Anopheles in the tropical Sub-Sahara Africa (SSA). However, the surveillance, monitoring and reporting of this epidemic is still poor and bureaucratic only. In our paper, we proposed a method to fast monitor and report Malaria instances by using Social Network Systems (SNS) and precipitation volume in Nigeria. We used Twitter search Application Programming Interface (API) to live-stream Twitter messages mentioning Malaria, preprocessed those Tweets and classified them into Malaria cases in Nigeria by using Support Vector Machine (SVM) classification algorithm and compared those Malaria cases with average precipitation volume. The comparison yielded a correlation of 0.75 between Malaria cases recorded by using Twitter and average precipitations in Nigeria. To ensure the certainty of our classification algorithm, we used an oversampling technique and eliminated the imbalance in our training Tweets.

• Journal of Korean Academy of Nursing
• /
• v.52 no.2
• /
• pp.144-156
• /
• 2022

Purpose: This study aimed to investigate the influence of social capital on the depression of older adults living in rural areas. Methods: Data sets were obtained from the 2019 Korea Community Health Survey. The participants were 39,390 older adults over 65 years old living in rural areas. Indicators of social capital included trust, reciprocity, network, and social participation. Depression-the dependent variable-was measured using the Patient Health Questionnaire-9 (PHQ-9). Hierarchical ordinal logistic regression was conducted to identify factors associated with depression after adjusting the data numbers to 102,601 by applying the Synthetic Minority Oversampling Technique (SMOTE). Results: The independent variables-indicators of social capital-exhibited significant association with the depression of older adults. The odds ratios of depression were higher in groups without social capital variables. Conclusion: To reduce depression, we recommend increasing social capital. Factors identified in this study need to be considered in older adult depression intervention programs and policies.