• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.49-55
• /
• 2020

Information theoretic learning (ITL) methods based on random symbols (RS) use a set of random symbols generated according to a target distribution and are designed nonparametrically to minimize the cost function of the Euclidian distance between the target distribution and the input distribution. One drawback of the learning method is that it can not utilize the input power statistics by employing a constant stepsize for updating the algorithm. In this paper, it is revealed that firstly, information potential input (IPI) plays a role of input in the cost function-derivative related with information potential output (IPO) and secondly, input itself does in the derivative related with information potential error (IPE). Based on these observations, it is proposed to normalize the step-size with the statistically varying power of the two different inputs, IPI and input itself. The proposed algorithm in an communication environment of impulsive noise and multipath fading shows that the performance of mean squared error (MSE) is lower by 4dB, and convergence speed is 2 times faster than the conventional methods without step-size normalization.

• Water for future
• /
• v.27 no.3
• /
• pp.45-54
• /
• 1994

Basic theory of fractal dimension is introduced and performed for the generated time series using the water balance model. The water balance equation over a large area is analyzed at seasonal time scales. In the generation and modification of mesoscale circulation local recycling of precipitation and dynamic effects of soil moisture are explicitly included. Time delay is incorporated in the analysis. Depending on the parameter values, the system showed different senarios in the evolution such as fixed point, limit cycle, and chaotic types of behavior. The stochastic behavior of the generated time series is due to deterministic chaos which arises from a nonlinear dynamic system with a limited number of equations whose trajectories are highly sensitive to initial conditions. The presence of noise arose from the characterization of the incoming precipitation, destroys the organized structure of the attractor. The existence of the attractor although noise is present is very important to the short-term prediction of the evolution. The implications of this nonlinear dynamics are important for the interpretation and modeling of hydrologic records and phenomena.

• Journal of Korean Society of Transportation
• /
• v.33 no.1
• /
• pp.14-28
• /
• 2015

Numerous research has been conducted using internet search data since the mid-2000s. For example, Google Inc. developed a service predicting influenza patterns using the internet search data. The main objective of this study is to prove the hypothesis that highway traffic indices are similar to the internet search patterns. In order to achieve this objective, a model to predict the number of vehicles entering the expressway and space-mean speed was developed and the goodness-of-fit of the model was assessed. The results revealed several findings. First, it was shown that the Google search traffic was a good predictor for the TCS entering traffic volume model at sites with frequent commute trips, and it had a negative correlation with the TCS entering traffic volume. Second, the Naver search traffic was utilized for the TCS entering traffic volume model at sites with numerous recreational trips, and it was positively correlated with the TCS entering traffic volume. Third, it was uncovered that the VDS speed had a negative relationship with the search traffic on the time series diagram. Lastly, it was concluded that the transfer function noise time series model showed the better goodness-of-fit compared to the other time series model. It is expected that "Big Data" from the internet search data can be extensively applied in the transportation field if the sources of search traffic, time difference and aggregation units are explored in the follow-up studies.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1999.05a
• /
• pp.425-431
• /
• 1999

We develope plasma monitoring system which detect plasma signals and store them for Laser welding quality evaluation and analysis using photo detector. The most fundamental and important aspects in such a system are signal restoration fidelity, noise immunity and noise cancelation capability. In this paper, we propose implementation method using distribute processing structure and hybrid digital communication for high noise cancelation capability, immunity and signal fidelity which are poorly presented in other researches. Lab experimental results and welding experimental results show a effectiveness of proposed method and plasma data is stored with 256 kbps without any communication error. we are implementing various welding defect recognition algorithm in this system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.11A
• /
• pp.1682-1690
• /
• 2000

최근들어 초광대역 시간 도약 임펄스 무선 전송(ultra wideband time hopping impulse radio; 이하 IR) 기술이 실내 무선 LAN이나 군용 무선 통신 시스템 등을위한 새로운대역 확산 전송 방식으로서 큰 관심을 모으고 있다 IR 시스템은 1(nsec) 이하의 매우 짧은 펄스폭의 펄스 위치 변조(pulse position modulation)된 가우시안 모노사이클 펄스열을 이용함으로서 매우작은 크기 의수 GHz에 이르는 초광대역 스펙트럼을 가지며 기존의 통신 시스템에 거의 간섭을 미치지 않으면서 사용 가능하다. 본 논문에서는 IR 시스템에 M진 (M-ary) PPM 방식을 적용하는 경우의 성능을 평가하였다. 특시 의사 잡음 부호 발생기와 10진 변환기(decimator)로 구성된 시간 도약 패턴을 사용하여 다원 접속 간섭이 배제되는 부가성 백색 가우시안 잡음 채널 하의 동기식(synchronous) IR 시스템에서, 심벌 수, M, 가우시안 모노사이클 펄스 폭 $\tau$$_{p}$, PPM에서 펄스 간 간격$\delta$, 펄스 반복 횟수 N$_{s}$ 등과 같은 다양한 시스템 파라미터의 조합에 따른 성능 변화를 모의 실험을 통해 확인하였으며, 이러한 결과를 통해 비트 오율, 시스템 복잡도 및 전송율 등의 사양에 따라 적합한 M 진 PPM IR 시스템의 구성 방안에 대한 근거를 제공하였다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.13 no.8
• /
• pp.797-803
• /
• 2002

In this paper we derive the probability density function of the phase error of the received signal over Rician fading channel and verify its propriety as the probability density function using the zeroth moment. In general, for the error probability over fading channel we compute the error probability in the first place when it is only AWGN, and then we get the final result by averaging the first result and the probability density function of the corresponding fading channel. In this paper, however, we compute the error probability by double integration after the probability density function over fading channel is computed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.635-638
• /
• 2005

The curve shape of the time fluctuation extracted from the electromagnetic signals is very complex. Thus it is important to decide exactly the signal property such as the inflection point for the observed signal. Usually filters elaborately designed are used to detect the signal characteristics. When the noise is added to the signal, the inflection point can be detected using the observation filter. In this paper we propose the design method for a natural observation filter with optimal filter order to extract a definite inflection point for the case of signals with the mixed noise.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.10
• /
• pp.717-727
• /
• 2022

An algorithm is realized for estimating the component fault and health diagnosis such as a deterioration. Based on the turbofan engine health diagnosis model, from the health parameters which are estimated by a real time tracking filter, the outliers are eliminated efficiently by an effective median filter to minimize an false alarm. The difference between the fault and deterioration trends is identified by the detection measure for abrupt change, thereby the clear diagnosis classifying the fault and the health condition is possible. The effectiveness of the algorithm for fault and health diagnosis is verified from the simulated results of engine component faults and deterioration.

• Journal of Internet Computing and Services
• /
• v.15 no.3
• /
• pp.71-77
• /
• 2014

The equalization algorithm based on the cross-information potential concept and Dirac-delta functions (CIPD) has outstanding ISI elimination performance even under impulsive noise environments. The main drawback of the CIPD algorithm is a heavy computational burden caused by the use of a block processing method for its weight update process. In this paper, for the purpose of reducing the computational complexity, a new method of the gradient calculation is proposed that can replace the double summation with a single summation for the weight update of the CIPD algorithm. In the simulation results, the proposed method produces the same gradient learning curves as the CIPD algorithm. Even under strong impulsive noise, the proposed method yields the same results while having significantly reduced computational complexity regardless of the number of block data, to which that of the e conventional algorithm is proportional.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.56-62
• /
• 2001

In this paper, we adopt a missing data theory to speech recognition. It can be used in order to maintain high performance of speech recognizer when the missing data occurs. In general, hidden Markov model (HMM) is used as a stochastic classifier for speech recognition task. Acoustic events are represented by continuous probability density function in continuous density HMM(CDHMM). The missing data theory has an advantage that can be easily applicable to this CDHMM. A marginalization method is used for processing missing data because it has small complexity and is easy to apply to automatic speech recognition (ASR). Also, a spectral subtraction is used for detecting missing data. If the difference between the energy of speech and that of background noise is below given threshold value, we determine that missing has occurred. We propose a new method that examines the reliability of detected missing data using voicing probability. The voicing probability is used to find voiced frames. It is used to process the missing data in voiced region that has more redundant information than consonants. The experimental results showed that our method improves performance than baseline system that uses spectral subtraction method only. In 452 words isolated word recognition experiment, the proposed method using the voicing probability reduced the average word error rate by 12％ in a typical noise situation.