• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.12
• /
• pp.1113-1119
• /
• 2000

To operate a process plant safely and economically, process monitoring is very important. Process monitoring is the task to identify the state of the system from sensor data. Process monitoring includes data acquisition, regulatory control, data reconciliation, fault detection, etc. This research focuses on the data recon-ciliation using scale-space filtering and fault detection using functional-link associative neural networks. Scale-space filtering is a multi-resolution signal analysis method. Scale-space filtering can extract highest frequency factors(noise) effectively. But scale-space filtering has too large calculation costs and end effect problems. This research reduces the calculation cost of scale-space filtering by applying the minimum limit to the gaussian kernel. And the end-effect that occurs at the end of the signal of the scale-space filtering is overcome by using extrapolation related with the clustering change detection method. Nonlinear principal component analysis methods using neural network have been reviewed and the separately expanded functional-link associative neural network is proposed for chemical process monitoring. The separately expanded functional-link associative neural network has better learning capabilities, generalization abilities and short learning time than the exiting-neural networks. Separately expanded functional-link associative neural network can express a statistical model similar to real process by expanding the input data separately. Combining the proposed methods-modified scale-space filtering and fault detection method using the separately expanded functional-link associative neural network-a process monitoring system is proposed in this research. the usefulness of the proposed method is proven by its application a boiler water supply unit.

• Journal of KIISE:Software and Applications
• /
• v.34 no.1
• /
• pp.30-39
• /
• 2007

The existing content-based spam mail filtering algorithms have difficulties in filtering spam mails when e-mails contain images but little text. In this thesis we propose an efficient spam mail classification algorithm that utilizes the link structure of e-mails. We compute the number of hyperlinks in an e-mail and the in-link frequencies of the web pages hyperlinked in the e-mail. Using these two features we classify spam mails and legitimate mails based on the decision tree trained for spam mail classification. We also suggest a hybrid system combining three different algorithms by majority voting: the link structure analysis algorithm, a modified link structure analysis algorithm, in which only the host part of the hyperlinked pages of an e-mail is used for link structure analysis, and the content-based method using SVM (support vector machines). The experimental results show that the link structure analysis algorithm slightly outperforms the existing content-based method with the accuracy of 94.8%. Moreover, the hybrid system achieves the accuracy of 97.0%, which is a significant performance improvement over the existing method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.11B
• /
• pp.722-729
• /
• 2005

Ethernet passive optical networks (EPONs) are an emerging access network technology which has a point-to-multipoint topology. EPONs operate point-to-multipoint in the OLT-ONU direction, and point-to-point in the ONU-OLT direction. To support point-to-point emulation and shared LAM emulation, EPONs use multi-point control protocol (MPCP). The MPCP uses logical link identification (LLID) field for frame tagging and filtering between the OLT and ONUs. In this paper, I propose logical-group identification (LGID) for logical link-based multicasting or VLAN services in EPONs. Using LGID with new frame tagging and filtering rules, EPONs support differentiated multimedia broadcasting or multicasting services. Additionally, EPONs can support logical link-based VLAN services that divides ONUs into several subsets.

• Korean Journal of Optics and Photonics
• /
• v.22 no.3
• /
• pp.122-128
• /
• 2011

The AWG-filtering effect was investigated on a bidirectional 100-GHz-channel-spacing WDM-PON link using spectrum-sliced and RSOA-amplified light sources for downstream signals and a wavelength reuse technique for upstream signals. Signal performances of three different filtering AWGs, including Gaussian, trapezoidal, and rectangular types, were compared on link transmission with fiber nonlinear effects. As an extinction ratio of a downstream signal varied, the effect for both directional signals was analyzed and optimized. It was found that there was an optimal pass bandwidth of an AWG for the balance between relative intensity noise decrement and cross phase modulation noise increment as the bandwidth got wider.

• Journal of Korea Spatial Information System Society
• /
• v.10 no.3
• /
• pp.55-65
• /
• 2008

For obtaining telematics traffic information(travel time or speed in an individual link), there are many kinds of devices to collect traffic data. Since the GPS satellite signals have been released to civil society, thank to the development of GPS technology, the GPS has become a very useful instrument for collecting traffic data. GPS can reduce the cost of installation and maintenance in contrast with existing traffic detectors which must be stationed on the ground. But. there are Problems when GPS data is applied to the existing filtering techniques used for analyzing the data collected by other detectors. This paper proposes a method to provide users with correct traffic information through filtering abnormal data caused by the unusual driving in collected data based on GPS. We have developed an algorithm that can be applied to real-time GPS data and create more reliable traffic information, by building patterns of past data and filtering abnormal data through selection of filtering areas using Quartile values. in order to verify the proposed algorithm, we experimented with actual traffic data that include probe cars equipped with a built-in GPS receiver which ran through Gangnam Street in Seoul. As a result of these experiments, it is shown that link travel speed data obtained from this algorithm is more accurate than those obtained by existing systems.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.3 no.1 s.4
• /
• pp.31-44
• /
• 2004

The study of Estimation model for the short-term travel time prediction. There is a different solution which has predicted the link travel time to solve this problem. By using this solution, the link travel time is predicted based on link conditions from time to time. The predicated link travel time is used to search the shortest path. Before providing a dynamic shortest path finding, the prediction model should be verified. To verify the prediction model, three models such as Kalman filtering, Stochastic Process, ARIMA. The ARIMA model should adjust optimal parameters according to the traffic conditions. It requires a frequent adjustment process of finding optimal parameters. As a result of these characteristics, It is difficult to use the ARIMA model as a prediction. Kalman Filtering model has a distinguished prediction capability. It is due to the modification of travel time predictive errors in the gaining matrix. As a result of these characteristics, the Kalman Filtering model is likely to have a non-accumulative errors in prediction. Stochastic Process model uses the historical patterns of travel time conditions on links. It if favorably comparable with the other models in the sense of the recurrent travel time condition prediction. As a result, for the travel time estimation, Kalman filtering model is the better estimation model for the short-term estimation, stochastic process is the better for the long-term estimation.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.6
• /
• pp.123-139
• /
• 2011

Web spam has a significant influence on the ranking quality of web search results because it promotes unimportant web pages. Therefore, web search engines need to filter web spam. web spam filtering is a concept that identifies spam pages - web pages contributing to web spam. TrustRank, Anti-TrustRank, Spam Mass, and Link Farm Spam are well-known web spam filtering algorithms in the research literature. The output of these algorithms depends upon the input seed. Thus, refinement in the input seed may lead to improvement in the quality of web spam filtering. In this paper, we propose seed refinement techniques for the four well-known spam filtering algorithms. Then, we modify algorithms, which we call modified spam filtering algorithms, by applying these techniques to the original ones. In addition, we propose a strategy to achieve better quality for web spam filtering. In this strategy, we consider the possibility that the modified algorithms may support one another if placed in appropriate succession. In the experiments we show the effect of seed refinement. For this goal, we first show that our modified algorithms outperform the respective original algorithms in terms of the quality of web spam filtering. Then, we show that the best succession significantly outperforms the best known original and the best modified algorithms by up to 1.38 times within typical value ranges of parameters in terms of recall while preserving precision.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.1
• /
• pp.108-120
• /
• 2017

To this end, the ROK military is developing a Korean Tactical Data Link (Link-K) in two stages: Basic and Completion types. Currently, the basic type of the Link-K has a limitations for supporting Mission Group(MG)s, such as to share unnecessary information and inefficient routing, because all RICCs simply broadcasts the same information in the basic type of the Link-K. In this paper, we propose a Zone-based Wireless Link-K network structure and a routing algorithm. To improve the effectiveness of the proposed routing algorithm, we have studied a filtering method that adjusts the message data according to the priority of the message and a method of controlling the load of the network. In order to evaluate the performance of the proposed scheme, it is proved that it is effective to support the mission task group compared with the existing wireless Link-K routing algorithm according to the operational scenario situation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.5D
• /
• pp.731-738
• /
• 2006

Facing congestion, people want to know traffic information about their routes, especially real-time link travel time (LTT). In this paper, as a sequel paper of the previous non-taxi based LTT generating study by Choi et al. (1998), taxi based GPS probes have been tried to produce LTT for urban arterials. Taxis in itself are good deployment mode of GPS probes although it by nature experiences boarding and alighting time noises which should be accounted. A heuristic real-time dynamic outlier filter algorithm for taxi GPS probe has been developed focusing on urban arterials. An actual traffic survey for dynamic link travel times has been conducted using license plate method for the test arterials of Seoul city transportation network. With the algorithm, it is estimated that 70% of outliers have been filtered and the relative error has been improved by 73.7%. The filtering algorithm developed here would be expected to be in use for other spatial sites with some calibration efforts. Some limitations and future research agenda have also been discussed.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.489-498
• /
• 2006

Clustering results differ according to the datasets and the performance worsens even while using web documents which are manually processed by an indexer, because although representative clusters for a feature can be obtained by statistical feature selection methods, irrelevant features(i.e., non-obvious features and those appearing in general documents) are not eliminated. Those irrelevant features should be eliminated for improving clustering performance. Therefore, this paper proposes three feature-filtering algorithms which consider feature values per document set, together with distribution, frequency, and weights of features per document set: (l) features filtering algorithm in a document (FFID), (2) features filtering algorithm in a document matrix (FFIM), and (3) a hybrid method combining both FFID and FFIM (HFF). We have tested the clustering performance by feature selection using term frequency and expand co link information, and by feature filtering using the above methods FFID, FFIM, HFF methods. According to the results of our experiments, HFF had the best performance, whereas FFIM performed better than FFID.