• Journal of Digital Policy
• /
• v.2 no.4
• /
• pp.25-32
• /
• 2023

This paper is a study to develop a deep neural network (DNN) blood glucose prediction model based on heart rate (HR) and heart rate variability (HRV) data measured by PPG-based sensors. MLP deep learning consists of an input layer, a hidden layer, and an output layer with 11 independent variables. The learning results of the blood glucose prediction model are MAE=0.3781, MSE=0.8518, and RMSE=0.9229, and the coefficient of determination (R²) is 0.9994. The study was able to verify the feasibility of glycemic control using non-blood vital signs using PPG-based digital devices. In conclusion, a standardized method of acquiring and interpreting PPG-based vital signs, a large data set for deep learning, and a study to demonstrate the accuracy of the method may provide convenience and an alternative method for blood glucose management in dogs.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.99-105
• /
• 2015

This paper deals with the mMMA (modified MMA) which possible to improving the convergence speed that employing the serial connecting form of two stage digital filter instead of signal filter of MMA adaptive equalizer without applying the variable step size for compensates the intersymbol interference by channel distortion in the nonconstant modulus signal. The adaptive equalizer can be implemented by signal digital filter using the finite order tap delay line. In this paper, the equalizer is implemented by the two stage serial form and the filter coefficient are updated by the error signal using the same algorithm of MMA in each stage. The fast convergence speed is determined in the first stage, and the residual isi left at the output of first stage output is minimized in the second stage filter. The same digital filter length was considered in single stage and two stage system and the performance of these systems were compared. The performance index includes the output signal constellation, the residual isi and maximum distortion, MSE that is measure of the convergence characteristics, the SER. As a result of computer simulation, mMMA that has a FIR structure of two stage, has more good performance in every performance index except the constellation diagram due to equalization noise and improves the convergence speed about 1.5~1.8 time than the present MMA that has a FIR structure of single stage.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.474-476
• /
• 2009

In the radio communication, such as echo cancellation and channel equalization, adaptive filtering is very practical. Its convergence behavior that is used for updating the weights depends on the correlation of the input signal and length of adaptive filter. Highly correlated input and long length of adaptive filter deteriorate the convergence behavior. To solve this problem, recently, subband affine projection algorithm which pre-whiten the correlation of the input and update the weights in subband structure has been presented. This paper presents convergence analysis method of multiple constrained subband affine projection algorithm.

• Journal of information and communication convergence engineering
• /
• v.9 no.2
• /
• pp.224-229
• /
• 2011

Cellular Automata is a discrete dynamical system that can be completely described in terms of local relation. For any given image, the system can save its features as well as increase or decrease the brightness of it locally through consideration of optimized transition in succession. These transitions in succession satisfy the function "Lyapunov" and have sequential movements. This study suggests the way of noise reduction for each image with the use of the Sequential Cellular Automata system. The mentioned transition in succession gives stable results with high-convergence performance to random noises and PSNR (Peak Signal-to-Noise Ratio) using histograms and MSE (Mean Square Error) for verification of effectiveness.

• Journal of electromagnetic engineering and science
• /
• v.4 no.1
• /
• pp.23-29
• /
• 2004

In this paper we analyze convergence constraints and time constant of IT-LMS algorithm and derive a method of making it's time constant independent of signal power by using input variance estimation. The method for estimating the input variance is to use a single-pole low-pass filter(LPF) with common smoothing parameter value, θ. The estimator is with narrow bandwidth for large θ but with wide bandwidth for small θ. This small θ gives long term average estimation(low frequency) of the fluctuating input variance well as short term variations (high frequency) of the input power. In our simulations of multipath communication channel equalization environments, the method with large θ has shown not as much improved convergence speed as the speed of the original IT-LMS algorithm. The proposed method with small θ=0.01 reach its minimum MSE in 100 samples whereas the IT-LMS converges in 200 samples. This shows the proposed, tap-power normalized IT-LMS algorithm can be applied more effectively to digital wireless communication systems.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.12a
• /
• pp.49-52
• /
• 2000

The Estimation method for spectral reflectance of the object using five-band and nine-band have been developed. The five-band acquisition are required of five or three times same work for color image acquisition process. To solve the above problems, we proposed a new method that can be reconstructed spectral reflectance of object. The proposed method was to classify same hues corresponding a color stimulus, by using hue angle and chroma vector of a color stimulus. The reconstruction of spectral reflectance was examined by computer simulation, and evaluated by MSE(Mean Square Error) and color difference between the original and reconstructed spectral reflectance.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.3135-3137
• /
• 1999

We propose two robust blind equalization algorithms based on multi-stage clustering blind equalization algorithm, which are called a complex classification update algorithm(CCUA) and an error compensation algorithm(ECA). The first algorithm is a tap-updating algorithm which each computes classified real and imaginary parts in order to reduce computations and the complexity of implementation as a stage increase. The second one is a algorithm which can achieve faster convergence speed because error of equalizer input make always fixed. Test results confirm that the proposed algorithms with faster convergence and lower complexity outperforms both constant modulus algorithm (CMA) and conventional multi-stage blind clustering algorithm(MSA) in reducing the SER as well as the MSE at the equalizer output.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.3
• /
• pp.63-69
• /
• 2013

This paper related with the performance of SE-MMA (Signed-Error MMA) that is the reduction of computational operation number in algorithm than MMA blind eualization algorithm which are possible to elimination of intersymbol interferance in the band limited and time dispersive nonlinear communication channel. In MMA algorithm which are possible to reduction of amplitude and phase rotation by intersymbol interference that is occurred in channel without using the training sequence, it uses the error signal that is the difference of the equalizer output and constant modulus, the statisticlly characteristic of transmitted signal. But in SE-MMA, it uses the polarity of the error signal, then it is possible to reduce the updating the tap coefficient and to simplify the H/W implementation. The computer simulation were performed in order to compare the performance of SE-MMA and conventional MMA algorithm. For this, the recovered signal constellation that is the output of the equalizer, the convergence performance by MSE, MD (maximum distortion) and residual isi characteristic learning curve, SER were used. As a result of simulation, the SE-MMA has more fast convergence speed than the MMA. But in the other index after reaching the seady state, it gives more worst performance values in the used index.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.1
• /
• pp.22-31
• /
• 2007

The proposed algorithm in this paper is based on the wavelet decomposition and the energy computation of composed blocks so the amount of calculation and complexity is minimized by adaptively replacing the DWT coefficients and managing the resources effectively. We are now living in the world of a lot. of multimedia applications for many digital electric appliances and mobile devices. Among so many multimedia applications, the digital image compression is very important technology for digital cameras to store and transmit digital images to other sites and JPEG2000 is one of the cutting edge technology to compress still images efficiently. The digital cm technology is mainly using the digital image compression features so that those images could be efficiently saved locally and transferred to other sites without any losses. JPEG2000 standard is applicable for processing the digital images usefully to keep, send and receive through wired and/or wireless networks. The discrete wavelet transform (DWT) is one of the main differences to the previous digital image compression standard such as JPEG, performing the DWT to the entire image rather than splitting into many blocks. Several digital images m tested with this method and restored to compare to the results of conventional DWT which shows that the proposed algorithm get the better result without any significant degradation in terms of MSE & PSNR and the number of zero coefficients when the energy based adaptive DWT is applied.

• Journal of the Institute of Convergence Signal Processing
• /
• v.18 no.2
• /
• pp.76-83
• /
• 2017

In this study, the original medical image, DICOM file, was converted into TIFF, BITMAP, GIF, JPEG image file, and then the conversion loss ratio according to the image compression and conversion process was quantitatively evaluated using Origin pro and ICY image analysis program. As the evaluation method, 50% MTF, structural similarity index, MSE, RMSE, maximum signal - to - noise ratio and so on were evaluated. The TIFF image file showed the same result as DICOM image in all experimental groups, Image file format. In this study, we propose a new method for evaluating the quality of digital images by applying original evaluation program such as Origin pro or ICY medical image analysis program. Is expected to be used as research data in the field of medical image processing, and TIFF image file showing the same result as DICOM image in the basic research field using digital medical image and evaluation program that does not support DICOM file Therefore, it is believed that it will help to secure reliability in digital medical image processing research using image file.