• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.3
• /
• pp.527-533
• /
• 2018

Recently, the device-to-device (D2D) communication has been conceived as the key technology for the next-generation mobile communication systems. The neighbor discovery in which the nearby users are found, is essential for the proper operation of the D2D communication. In this paper, we propose new neighbor discovery scheme based on deep learning technology which has gained a lot of attention recently. In the proposed scheme, the neighboring users can be found using the uplink pilot transmission of users only, unlike conventional neighbor discovery schemes in which direct pilot communication among users is required, such that the signaling overhead can be greatly reduced in our proposed scheme. Moreover, the neighbors with different proximity can also be classified accordingly which enables more accurate neighbor discovery compared to the conventional schemes. The performance of our proposed scheme is verified through the tensorflow-based computer simulations.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.307-332
• /
• 2022

One of the most intensively conducted research areas in business application study is a bankruptcy prediction model, a representative classification problem related to loan lending, investment decision making, and profitability to financial institutions. Many research demonstrated outstanding performance for bankruptcy prediction models using artificial intelligence techniques. However, since most machine learning algorithms are "black-box," AI has been identified as a prominent research topic for providing users with an explanation. Although there are many different approaches for explanations, this study focuses on explaining a bankruptcy prediction model using a counterfactual example. Users can obtain desired output from the model by using a counterfactual-based explanation, which provides an alternative case. This study introduces a counterfactual generation technique based on a genetic algorithm (GA) that leverages both domain knowledge (i.e., causal feasibility) and feature importance from a black-box model along with other critical counterfactual variables, including proximity, distribution, and sparsity. The proposed method was evaluated quantitatively and qualitatively to measure the quality and the validity.

• Journal of Navigation and Port Research
• /
• v.45 no.6
• /
• pp.306-313
• /
• 2021

National fishing harbour is as a refuge for fishing boats and a breakwater lighthouse is installed as a functional facility in consideration of harbour identification and the safety of passing vessels. In this study, the service level of breakwater lighthouse (234 units) was estimated based on the traffic volume of 105 national fishing harbour. For this purpose, the evaluation items were determined, the fishing harbour standard index was calculated (Fs=1), the proximity of fishing harbour was identified and the function (service level) of the breakwater lighthouse was estimated in the following order. However, national fishing harbour differed in size, traffic volume and fishing vessel capacity. Therefore, 105 national fishing harbour were divided into three groups through cluster analysis. The cluster analysis was based on the service level factors of the breakwater lighthouse, such as the number of weeding fishing vessels, tonnage of fishing vessels, the number of incoming and outgoing vessels per year, and fishing vessel capacity. As a result of the estimation, the service level of the breakwater lighthouse (light tower height, visual height, visual range, interval) was 10.50m, 16.50m, 7.00mile, 5.5sec for group 1, and 10.67m, 16.16m, 8.33mile, and 6.0sec for group 2, The three groups are 11.53m, 16.75m, 6.75mile and 5.0sec. The results of this study can be used as useful basic data for improving the service level of traffic vessels when a breakwater lighthouse is built in a fishing harbour in the future.

• Journal of the Korean Geographical Society
• /
• v.38 no.3
• /
• pp.426-443
• /
• 2003

Electronic Commerce (EC) has been at the center of discussion as a symbol of the integration of unprecedentedly developed Information and Communication Technologies (ICTs) and traditional commerce. In spite of much attention to EC, the research from a spatial perspective has not proliferated yet. EC was regarded to have aspatial characteristics based on the expectation for a global expansion of business activities in a digital economy. This paper attempts to figure out the spatial characteristics of public Business-to-Business electronic marketplaces (public B2B e-MPs), as one of the most evolving forms of B2B EC, regardless of the low proportion in B2B EC at present. Many of the firms operating public B2B e-MPs in Korea were located in Seoul, especially in Gangnam-gu. The analysis of three spatial indices showed their extreme spatial concentration. The analysis on the location factors of Public B2B e-MP firms in Korea demonstrated that location factors of public B2B e-MP firms were differentiated by regional groups: Gangnam-gu, Seoul except Gangnam-gu, and the provinces. It was againt an initial extreme expectation that the firms relevant to B2B EC will not care about physical locations because they mainly do businesses in electronic space. The differences between those in Gangnam-gu and in the provinces were strikingly prominent. Such differentiated location factors by region were closely related to the different attributes of the public B2B e-MPs by region. In conclusion, public B2B e-MPs are not irrelevant to physical space and physical proximity, at least at a current stage. Customized spatial strategies are required for successful online businesses.

• Journal of dental hygiene science
• /
• v.13 no.2
• /
• pp.174-181
• /
• 2013

The objective of this research is to compare and analyze regional accessibility of Korean primary school students to oral health services from the perspective of public health geography by using geographic information system in which the choropleth map has been regarded as the most popular method. Statistical proximity on the basis of calculus of 205 regions-based school oral health data is optimized to set five class intervals for five maps. These choropleth maps of oral heal programs such as oral health education, tooth-brushing method education, preventive dental care and curative dental care, demonstrate that there exist wide regional discrepancies throughout the country in terms of primary school students' accessibility to oral health services within the programs. The paper not just contributes to overcoming the existing paradigm by actively considering an interdisciplinary research among public health dentistry, dental hygiene and geography of public health, but provides clear evidence for national oral health policy in South Korea.

• Journal of Broadcast Engineering
• /
• v.21 no.3
• /
• pp.425-435
• /
• 2016

In this paper, we propose a classification model by analyzing raw material images recorded using a color CCD camera to automatically classify good and defective agricultural products such as rice, coffee, and green tea, and raw materials. The current classifying agricultural products mainly depends on visual selection by skilled laborers. However, classification ability may drop owing to repeated labor for a long period of time. To resolve the problems of existing human dependant commercial products, we propose a vision based automatic raw material classification combining mean shift clustering and stepwise region merging algorithm. In this paper, the image is divided into N cluster regions by applying the mean-shift clustering algorithm to the foreground map image. Second, the representative regions among the N cluster regions are selected and stepwise region-merging method is applied to integrate similar cluster regions by comparing both color and positional proximity to neighboring regions. The merged raw material objects thereby are expressed in a 2D color distribution of RG, GB, and BR. Third, a threshold is used to detect good and defective products based on color distribution ellipse for merged material objects. From the results of carrying out an experiment with diverse raw material images using the proposed method, less artificial manipulation by the user is required compared to existing clustering and commercial methods, and classification accuracy on raw materials is improved.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.627-633
• /
• 2020

Assembly process data can be gathered in real time in a manufacturing execution system (MES) server using proximity sensors, barcodes, RFID, ZigBee, Bluetooth, wireless sensor networks, etc. Although this is suitable for identifying process flow and checking production progress, it is difficult to trace the location of individual workers in real time for missing work or trajectories within the work area. To overcome this, the location and trajectory of the working tool can be analyzed in real time through a position tracking system of an operator's working tool. It can instruct the operator to perform a consistent working process. Productivity and quality improvement can be achieved by an alarming or blocking operator with possible assembly defects during the assembly process in real time. To this end, we developed a real-time tool position-tracking sensor system based on Ultra Wide Band (UWB) trilateration using a Kalman filter to eliminate mechanical vibration and radio communication noise.

• The Journal of the Korea Contents Association
• /
• v.16 no.10
• /
• pp.588-597
• /
• 2016

According to an annual industrial accident report from Ministry of Employment of Labor, among the various types of accidents, the number of accidents from construction industry increases every year with the percentage of 27.56% as of 2014. In fact, this number has risen almost 3% over the last four years. Currently, among the industrial accidents, heavy machinery causes most of the tragedy such as collision or narrowness. As reported by the government, most of the time, both heavy machinery drivers and workers were unaware of each other's positions. Nowadays, however when society requires highly complex structures in minimal time, it is inevitable to allow heavy construction equipments running simultaneously in a construction field. In this paper, we have developed Approach Detection System for excavator in order to reduce the increasing number. The imaged based Approach Detection System contains camera, approach detection sensor and Around View Monitor (AVM). This system is also applicable in a small scale construction fields along with other machineries besides excavators since this system does not require additional communication infra such as server.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.3
• /
• pp.620-628
• /
• 2017

In this paper, we suggest a route plan sharing service for the navigation assistance service which is the second item of 16 items in maritime service portfolios (MSPs) for safety navigation based on interview process. Also, we developed scenarios for effectiveness evaluation of the proposed service, and conducted simulations using full mission ship handling simulator (FMSS) for effectiveness evaluation of proposed services based on the developed scenarios. Through the simulations, we analyzed proximity measures, controllability statistics and subjective evaluations to assess the usefulness of suggested service. If accomplishing the test for new services to apply real ships and vessel traffic (VTS) center, there has possibilities to occur various risks in terms of time/cost problems. Therefore, there is needs for the preliminary effectiveness evaluation processes necessarily when adopts and implements new services. Because we expected the service that is helpful for safety navigation, but the test results are not when conducted a simulation.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.4
• /
• pp.53-64
• /
• 2009

Inter-vehicular communication that propagates information without infrastructures has drawn a lot of interest. However, it is difficult to apply conventional ad-hoc routing protocols directly in inter-vehicular communication due to frequent changes in the network topology caused by high mobility of the vehicles. MMFP(Multi-hop MAC Forwarding) is a unicast forwarding protocol that transport packets based on the reachability information instead of path selection or position information. However, delivering public safety messages informing road conditions such as collision, obstacles and fog through inter-vehicular communication requires broadcast rather than unicast since these messages contain information valuable to most drivers within a close proximity. Flooding is one of the simplest methods for multi-hop broadcast, but it suffers from reduced packet delivery-ratio and high transmission delay due to an excessive number of duplicated packets. This paper presents two multi-hop broadcast protocols for inter-vehicular communication that extend the MMFP. UMHB(Unreliable Multi-Hop Broadcast) mitigates the duplicated packets of MMFP by limiting the number of nodes to rebroadcast packets. UMHB, however, still suffers from low delivery ratio. RMHB(Reliable Multi-Hop Broadcast) uses acknowledgement and retransmission in order to improve the reliability of UMHB at the cost of increase in transmission delay, which we show through simulation is within an acceptable range for collision avoidance application.