• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• The KIPS Transactions:PartC
• /
• v.17C no.6
• /
• pp.441-450
• /
• 2010

Recently, a lot of interest is increased in sensor network which gathers various data through many sensor nodes deployed in wired and wireless network environment. However, because of the limitation in memory, computation, and energy of the sensor nodes, security problem is very important issue. In sensor network, not only the security problem, but also computing power should be seriously considered. In this paper, considering these characteristics, we make the sensor network consist of normal sensor nodes and clusterheaders with enough space and computing power, and propose a group key rekeying scheme adopting PCGR(Predistribution and local Collaborationbased Group Rekeying) for secure group communication. In our proposal, we enhance the security by minimizing the risk to safety of the entire network through verifying the new key value from clusterheader by sensor nodes. That is, to update the group keys, clusterheaders confirm sensor nodes through verifying the information from sensor nodes and send the new group keys back to authentic member nodes. The group keys sent back by the clusterheaders are verified again by sensor nodes. Through this mutual authentication, we can check if clusterheaders are compromised or not. Qualnet simulation result shows that our scheme not only guarantees secure group key rekeying but also decreasesstorage and communication overhead.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.2
• /
• pp.177-187
• /
• 2000

In the past several years, many systems which adopted ring topology with high-speed unidirectional point-to-point links have emerged to overcome the limit of bus for interconnection network of clustered multiprocessor system. However, rapid increase of processor speed and performance improvement of local bus and memory system limit scalability of system with point-to-point link of standard bandwidth. Therefore, necessity to extend bandwidth is emphasized. In this paper, we adopt PANDA system as base model, which is clustering-based multiprocessor system. By simulating a model adopting commercial processor and local bus specification, we show that point-to-point link is bottleneck of system performance, and bandwidth expansion by more than 200% is needed. To expand bandwidth of interconnection network, it needs excessive design cost and time to develop new point-to-point link with doubled bandwidth. As an alternative to double bandwidth, we propose several ways to implement dual ring -simple dual ring, transaction-separated dual ring, direction-separated dual ring- by using off-the-shelf point-to-point links with IEEE standard bandwidth. We analyze pros. and cons. of each model compared with doubled-bandwidth single ring by simulation.

• Archives of Plastic Surgery
• /
• v.39 no.5
• /
• pp.540-545
• /
• 2012

Background Patients have anxiety and fear of complications due to general anesthesia. Through new instruments and local anesthetic drugs, a variety of anesthetic methods have been introduced. These methods keep hospital costs down and save time for patients. In particular, the target-controlled infusion (TCI) system maintains a relatively accurate level of plasma concentration, so the depth of anesthesia can be adjusted more easily. We conducted this study to examine whether intravenous anesthesia using the TCI system with propofol and remifentanil would be an effective method of anesthesia in breast augmentation. Methods This study recruited 100 patients who underwent breast augmentation surgery from February to August 2011. Intravenous anesthesia was performed with 10 mg/mL propofol and 50 ${\mu}g/mL$ remifentanil simultaneously administered using two separate modules of a continuous computer-assisted TCI system. The average target concentration was set at 2 ${\mu}g/mL$ and 2 ng/mL for propofol and remifentanil, respectively, and titrated against clinical effect and vital signs. Oxygen saturation, electrocardiography, and respiratory status were continuously measured during surgery. Blood pressure was measured at 5-minute intervals. Information collected includes total duration of surgery, dose of drugs administered during surgery, memory about surgery, and side effects. Results Intraoperatively, there was transient hypotension in two cases and hypoxia in three cases. However, there were no serious complications due to anesthesia such as respiratory difficulty, deep vein thrombosis, or malignant hypertension, for which an endotracheal intubation or reversal agent would have been needed. All the patients were discharged on the day of surgery and able to ambulate normally. Conclusions Our results indicate that anesthetic methods, where the TCI of propofol and remifentanil is used, might replace general anesthesia with endotracheal intubation in breast augmentation surgery.

• Journal of Digital Contents Society
• /
• v.12 no.1
• /
• pp.33-48
• /
• 2011

In this thesis, a field experience activity support system is designed and implemented for improving social skills of children with developmental disabilities. The proposed system is designed to allow the students to experience various field activities around their local community. The proposed system has the following characteristics. First, it can improve effectiveness of field experience by providing very practical pre-study information to students. Second, the systems provides very practical and real-world problem-solving abilities rather than providing simple experience and superficial information. Third, the system allows students to change the contents according to their local community environments and purpose of use. Fourth, the system can be used as education for living at home as well as field experience study at schools. After applying the proposed system to students with developmental disabilities, the following positive results are obtained. First, the system have good effect on students with severe developmental disabilities when the system is used for prestudy. Specially those students are highly interested and motivated on study subjects. Second, at real field experience study place, students are well adapted and are very interested in their activities. Third, in the light of post-evaluation after field experience, it is reported that lots of study contents remain in their memory.

• Journal of Advanced Marine Engineering and Technology
• /
• v.39 no.4
• /
• pp.483-488
• /
• 2015

This paper implements a long-term monitoring system (LMS) for ship's engine performance analysis (SEPA) based on the web, for the purpose of the communication speed and engine maintenance. This system is composed of a simulator, monitoring module with a multi channel A/D converter, monitoring computer, network attached storage (NAS), RS485 serial and wireless internet communication system. The existing products monitor the information transmitted from pressure sensors installed in the upper parts of each of engines in the local or web computer, but have a delay in the communication speed and errors in long-term monitoring due to the large volume of sampling pressure data. To improve these problems, the monitoring computer saves the sampling pressure data received from the pressure sensors in NAS, monitors the long-term sampling data generated by the sectional down sampling method on a local computer, and transmits them to the web for long-term monitoring. Because this method has one tenth of the original sampling data, it will use memory with small capacity, save communication cost, monitor the long-term sampling data for 30 days, and as a result, make a great contribution to engine maintenance.

• Smart Structures and Systems
• /
• v.11 no.5
• /
• pp.555-567
• /
• 2013

An Remote Corrosion Monitoring (RCM) system consists of an anode with low potential, the metallic structures against corrosion, an electrode to provide reference potential, and a data-acquisition system to ensure the potential difference for anticorrosion. In more detail, the data-acquisition (DAQ) system monitors the potential difference between the metallic structures and a reference electrode to identify the correct potential level against the corrosion of the infrastructures. Then, the measured data are transmitted to a central office to remotely keep track of the status of the corrosion monitoring (CM) system. To date, the RCM system is designed to achieve low power consumption, so that it can be simply powered by batteries. However, due to memory effect and the limited number of recharge cycles, it can entail the maintenance fee or sometimes cause failure to protect the metallic structures. To address this issue, the low-overhead energy harvesting circuitry for the RCM systems has designed to replenish energy storage elements (ESEs) along with redeeming the leakage of supercapacitors. Our developed energy harvester can scavenge the ambient energy from the corrosion monitoring environments and store it as useful electrical energy for powering local data-acquisition systems. In particular, this paper considers the energy harvesting from potential difference due to galvanic corrosion between a metallic infrastructure and a permanent copper/copper sulfate reference electrode. In addition, supercapacitors are adopted as an ESE to compensate for or overcome the limitations of batteries. Experimental results show that our proposed harvesting schemes significantly reduce the overhead of the charging circuitry, which enable fully charging up to a 350-F supercapacitor under the low corrosion power of 3 mW (i.e., 1 V/3 mA).

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.4
• /
• pp.144-154
• /
• 2013

This paper proposes a new super-resolution algorithm where sharpness enhancement is merged in order to improve overall visual quality of up-scaled images. In the learning stage, multiple dictionaries are generated according to sharpness strength, and a proper dictionary among those dictionaries is selected to adapt to each patch in the inference stage. Also, additional post-processing suppresses boosting of artifacts in input low-resolution images during the inference stage. Experimental results that the proposed algorithm provides 0.3 higher CPBD than the bi-cubic and 0.1 higher CPBD than Song's and Fan's algorithms. Also, we can observe that the proposed algorithm shows better quality in textures and edges than the previous works. Finally, the proposed algorithm has a merit in terms of computational complexity because it requires the memory of only 17% in comparison with the previous work.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.7
• /
• pp.515-519
• /
• 2009

As the application fields of sensor nodes are getting diverse these days, it is required to have a way of collecting various data that is suitable for these application fields. In the case that the real-time surveillance over the data is unnecessary, present data collecting methods, which collect and transfer the data directly, can cause a waste of energy and data loss, A new method that store the collected data in a local storage and acquire them by query later on is required for nonreal-time applications. NAND flash has energy efficiency and large capacity so it is suitable for sensor nodes, Sensor nodes support 4-10 KBytes small sized memory and it is hard to build an effective file system since NAND Flash doesn't support overwriting NAND flash. This paper discusses an implementation of NAND Flash file system in sensor node environments. The file system makes long-term data collecting possible by reducing transmission cost. It is expected that this file system will play a central role in sensor network environments as it can be applied to various fields which call for long term data collecting.

• Information Systems Review
• /
• v.5 no.1
• /
• pp.97-111
• /
• 2003

With the spread of electronic commerce on the Internet, comparison shopping sites with agent technique are getting popular for the best shopping. However, most consumers are still spending much time to search for the best price through the sites because each of them may show a different price even for the same goods or a site does not show any information about specific goods. Additionally, the search for the best price of the goods like books and CDs may cause the system to be overloaded and the response time to be long due to an on-line real-time search. In this paper, we have designed and implemented a meta-search system for comparison shopping sites with a local database and memory cache to resolve the above problems. The proposed system collects and maintains the price information of popular goods among the comparison shopping sites using several software agents. The experimental results show that our system is an efficient meta-comparison shopping engine and reduces the latency of the response time with little overhead.