• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.05a
• /
• pp.382-388
• /
• 2007

In this paper, a modified Fuzzy C-Means (MFCM) algorithm is presented for nonlinear blind channel equalization. The proposed MFCM searches the optimal channel output states of a nonlinear channel from the received symbols, based on the Bayesian likelihood fitness function instead of a conventional Euclidean distance measure. Next, the desired channel states of a nonlinear channel are constructed with the elements of estimated channel output states, and placed at the center of a Radial Basis Function (RBF) equalizer to reconstruct transmitted symbols. In the simulations, binary signals are generated at random with Gaussian noise. The performance of the proposed method is compared with that of a hybrid genetic algorithm (GA merged with simulated annealing (SA): GASA), and the relatively high accuracy and fast searching speed are achieved.

• The KIPS Transactions:PartC
• /
• v.12C no.1 s.97
• /
• pp.63-68
• /
• 2005

In this paper, performance of a maximum-likelihood modulation classification for quadrature amplitude modulation (QAM) is studied. Unlike previous works, the relative classification performance with respect to the available modulations and performance limit with single-sample observation are presented. For those purposes, all constellations are set to have the same minimum Euclidean distance between symbols so that a smaller constellation is a subset of the larger ones. And only one sample of received waveform is used for multiple hypothesis test. As a result, classification performance is improved with increase in signal-to-noise ratio in all the experiments. Especially, when the true modulation format used in the transmitter is 4 QAM, almost perfect classification can be achieved without any additional information or observation samples. Though the possibility of false classification due to the symbols shared by subset constellations always exists, correct classification ratio of $80{\%}$ can be obtained with the single-sample observation when the true modulation formats are 16 and 64 QAM.

• The Korean Journal of Applied Statistics
• /
• v.16 no.2
• /
• pp.455-467
• /
• 2003

There has been a long debate on the applicability of the chi-square approximation to statistics based on small sample size. Extending comparison results among Pearson chi-square Χ$^2$, generalized likelihood .ratio G$^2$, and the power divergence Ι(2/3) statistics suggested by Rudas(1986), recently developed disparity statistics (BWHD(1/9), BWCS(1/3), NED(4/3)) we compared and analyzed in this paper. By Monte Carlo studies about the independence model of two dimension contingency tables, the conditional model and one variable independence model of three dimensional tables, simulated 90 and 95 percentage points and approximate 95% confidence intervals for the true percentage points are obtained. It is found that the Χ$^2$, Ι(2/3), BWHD(1/9) test statistics have very similar behavior and there seem to be applcable for small sample sizes than others.

• Journal of Life Science
• /
• v.14 no.4
• /
• pp.593-607
• /
• 2004

The nucleotide sequence for a nuclear-encoded small subunit rDNA (SSU rDNA) was determined for 43 species of the class Dinophyceae, including harmful algae Cochlodinium polykrikoides and Gyrodinium aureolum. These sequences and data analyses were performed by parsimony, distances and maximum likelihood methods in PHYLIP (Phylogenetic Inference Package) version 3.573c. The species Noctiluca scintillans, Gonyaulax spinifern and Crypthecodinium cohnii occupied a basal position within the Dino- phyceae in our analyses. The genera Alexandrium and Symbiodinium were monophyletic (supported by a bootstrap value of >70%), whereas the genera Gymnedinium and Gyrodinium formed polyphyletic nodes, for which bootstrap support was strong (>70%) in the neighbor-joining and maximum likelihood methods except for the PHYLIP parsimony analysis (=59%). The sequence divergence between G. aureolum and G. dorsum/ G. galathenum was the largest at 7.4% (45 bp), whereas G. aureolum and G. mikimotoi showed an extremely low value of genetic divergence of 0.9% (5 bp). The genetic divergence between C. polykrikoides and G. aureolum was a low value of 5.2% (31 bp). In the phylogenetic analysis, the placement of G. aureolum and C. polykrikoides was closer to the genus Gymnodinium than to the genus Gyrodinium, which was supported by a moderate bootstrap value.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.3A
• /
• pp.223-231
• /
• 2011

It has been shown that the performance of multiple-input multiple-output (MIMO) spatial multiplexing systems is significantly degraded when spatial correlation exists between transmit and receive antenna pairs. In this paper, we investigate designs of a new statistical precoder for spatial multiplexing systems with maximum likelihood (ML) receiver which requires only correlation statistics at the transmitter. Two kinds of closed-form solution precoders based on rotation and power allocation are proposed by means of maximizing the minimum E tlidean distance of joint symbol constellations. In addition, we extend our results to linear receivers for correlated channels. We provide a method which yields the same profits from the proposed precoders based on a simple zero-forcing (ZF) receiver. The simulation shows that 2dB and 8dB gains are achieved for ML and ZF systems with two transmit antennas, respectively, compared to the conventional systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5A
• /
• pp.375-387
• /
• 2012

This paper proposes a new real-time object tracking algorithm using particle filters with color and texture features in moving CCD camera images. If the user selects an initial object, this region is declared as a target particle and an initial state is modeled. Then, N particles are generated based on random distribution and CS-LBP (Centre Symmetric Local Binary Patterns) for texture model and weighted color distribution is modeled from each particle. For observation likelihoods estimation, Bhattacharyya distance between particles and their feature models are calculated and this observation likelihoods are used for weights of individual particles. After weights estimation, a new particle which has the maximum weight is selected and new particles are re-sampled using the maximum particle. For performance comparison, we tested a few combinations of features and particle filters. The proposed algorithm showed best object tracking performance when we used color and texture model simultaneously for likelihood estimation.

• Journal of Satellite, Information and Communications
• /
• v.9 no.2
• /
• pp.23-27
• /
• 2014

In this paper, we propose a soft demapping method for 64-ary APSK in the DVB-S3 system. The proposed method in this paper uses the hard decision threshold (HDT) line for each constituent bit in a symbol, and calculates the soft bit information with the distance between the HDT line and the detected symbol. If the HDT lines are defined in a simple manner, the complexity to estimate soft information can be largely reduced compared with the maximum likelihood detection (MLD) which has an exponential complexity. By considering this, we first derive HDT lines for each constituent bit for a 64-APSK symbol, and propose a method to calculate soft bit information. We simulate the BER performance of the proposed scheme by using a turbo codes which requires soft-input-soft-output information, and compare it that of the MLD. The result show that the proposed scheme produces approximating performance to MLD with largely reduced complexity.

• Korean Journal of Remote Sensing
• /
• v.13 no.1
• /
• pp.47-56
• /
• 1997

In this paper, remote sensing is used to estimate trophic state which is primary concern in a lake. In using remote sensing, this study estimated trophic state not with conventional method such as regression equations but with classification methods. As europhication is caused by the extraodinary proliferation of the algae, chlorophyll a and transparency are applied to remote sensing data.. Maximum Likelihood Classification and Minimum Distance Classification which are kinds of classification methods enabled trophic state to be confirmed in a lake. These are obtained as the result of applying remote sensing to classify trophic state in a lake. Firest, when we evaluate tropic state in a large area of water body, the application of remote sensing data can obtain more than 70% accuracies just in using basic classification methods. Second, in the aspect of classification, the accuracy of Minimum Distance Classification is usually better than that of Maximum Likelihood Classification. This result is caused that samples have normal distribution, but their numbers are a few to apply statistical method. Therefore, classification method is required such as artificial neural networks which are not influenced by statistical distribution. Third, this study enables the trophic state of water body to be analyzed and evaluated rapidly, periodically and visibly. Also, this study is good for forming proper countermeasure accompanying with trophic state progress extent in a lake and is useful for basic-data.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.6
• /
• pp.552-558
• /
• 2002

Recently, based on the human speech recognition (HSR) model of Fletcher, the multi-band speech recognition has been intensively studied by many researchers. As a new automatic speech recognition (ASR) technique, the multi-band speech recognition splits the frequency domain into several sub-bands and recognizes each sub-band independently. The likelihood scores of sub-bands are weighted according to reliabilities of sub-bands and re-combined to make a final decision. This approach is known to be robust under noisy environments. When the noise is stationary a sub-band SNR can be estimated using the noise information in non-speech interval. However, if the noise is non-stationary it is not feasible to obtain the sub-band SNR. This paper proposes the inverse sub-band distance (ISD) weighting, where a distance of each sub-band is calculated by a stochastic matching of input feature vectors and hidden Markov models. The inverse distance is used as a sub-band weight. Experiments on 1500∼1800㎐ band-limited white noise and classical guitar sound revealed that the proposed method could represent the sub-band reliability effectively and improve the performance under both stationary and non-stationary band-limited noise environments.

• Journal of Distribution Science
• /
• v.12 no.6
• /
• pp.67-79
• /
• 2014

Purpose - This study aims to accomplish three major research goals. First, it strives to change consumers' focus from peripheral routes to a central route of public service advertising related to the good health policy, without problematic effects, by influencing consumers' knowledge or involvement. Second, this study examines the elaboration likelihood model (ELM) and construal level theory (CLT). Specifically, we consider that the central route of ELM might correspond with the focal goal of CLT. Third, this study analyzes ELM through CLT. That is, ELM predicted that low involvement would take the peripheral route, and high involvement would take the central route. Research design, data, and methodology - This study consisted of three experiments. The first experiment had a 2×2 between-subject design. The subjects were university students and the research period was approximately one year. The first independent variable was the involvement of the overweight issue; this variable was measured and split by the median. The second independent variable was the temporal distance (near vs. distant future); this variable was manipulated. The second experiment also had a 2×2 between-subject design. The first variable was the involvement of cervical adenocarcinoma prevention, and was considered already manipulated by sex. Specifically, males had a low involvement of the disease, but females had high involvement. The second independent variable was priming (power vs. submissive). Power priming would induce abstract thinking, but submissive priming would take concrete processing. The third experiment had a 2×2×2 between-subject design. The first variable was cognitive depletion, and was manipulated by memorizing 9-digit numbers. The second and third independent variables were involvement and abstract thinking induction, such as prior experiments. Data were collected through questionnaires, and were analyzed by an SPSS program. Major hypotheses were tested by examining the interaction effects through ANOVA. Results - Major findings are as follows. First, even for low-involved consumers in the overweight category, distant future manipulation induced them to focus not on the peripheral route but on the central route of the public service advertisement. This result does not correspond to the typical ELM prediction. Second, under power priming, low-involved males of the cervical adenocarcinoma category focused on the peripheral route because of the induction to abstract thinking. This result replicated the first experiment, and confirmed the theoretical robustness. Third, high-involved females focused not on the central but on the peripheral route under the mixed condition of cognitive depletion and near future manipulation. Depletion consumed cognitive resources, and the processing mode of consumers changed from systematic to heuristic. Conclusions - ELM needs to be complemented through CLT in context of public service good health advertising. Specifically, the involvement of ELM may impact consumers' thinking mode (abstract vs. concrete), and the interaction effects may influence consumers' focus on advertising (central vs. peripheral route). This study's limitations were bounded subjects, limited stimuli, and somewhat weak external validity.