• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.10
• /
• pp.941-948
• /
• 2016

In this paper, we efficiently simulated for collecting the data from the fixed sensor and mobile sensor of patients using the LEACH-Mobile method. The LEACH-Mobile method is the protocol to increase the mobility by adding the mobile node to the existed LEACH(:Low Energy Adaptive Clustering Hierarchy) protocol. It improves the mobility of The LEACH-Mobile in the LEACH, however it consumes more energy than the existed LEACH. There is reason why we use the LEACH-Mobile that the monitoring system is generally done by the CCTV and an periodic checkup by nurses. However the number of nurse is a few in the most of hospital. It can happen the accidents of the patients in out of the CCTV when the nurse can not see the monitoring system in the hospital. Therefore it is simulated to continuously gather the data of the position and sensors in the five situation of moving the patients in the hospital, it gets the result that the management of the mobile patients is more efficient.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.2
• /
• pp.66-77
• /
• 2011

We propose a framework for grouping photographic advertising images that employs a hierarchical indexing scheme based on efficient feature combinations. The study provides one specific application of effective tools for monitoring photographic advertising information through online and offline channels. Specifically, it develops a preprocessor for advertising image information tracking. We consider both global features that contain general information on the overall image and local features that are based on local image characteristics. The developed local features are invariant under image rotation and scale, the addition of noise, and change in illumination. Thus, they successfully achieve reliable matching between different views of a scene across affine transformations and exhibit high accuracy in the search for matched pairs of identical images. The method works with global features in advance to organize coarse clusters that consist of several image groups among the image data and then executes fine matching with local features within each cluster to construct elaborate clusters that are separated by identical image groups. In order to decrease the computational time, we apply a conventional clustering method to group images together that are similar in their global characteristics in order to overcome the drawback of excessive time for fine matching time by using local features between identical images.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1397-1404
• /
• 2018

Due to the technical difficulties and high cost for obtaining 3-dimensional structure data, sequence-based approaches in proteins have not been widely acknowledged. A motif can be defined as any segments in protein or gene sequences. With this simplicity, motifs have been actively and widely used in various areas. However, the motif itself has not been studied comprehensively. The value of this study can be categorized in three fields in order to analyze the human proteins using artificial intelligence method: (1) Based on our best knowledge, this research is the first comprehensive motif analysis by analyzing motifs with all human proteins in Protein Data Bank (PDB) associated with the database of Enzyme Commission (EC) number and Structural Classification of Proteins (SCOP). (2) We deeply analyze the motif in three different categories: pattern, statistical, and functional analysis of clusters. (3) At the last and most importantly, we proposed random motif frequency(RMF) matric that can efficiently distinct the characteristics of proteins by identifying interface residues from non-interface residues and clustering protein functions based on big data while varying the size of random motif.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.6
• /
• pp.701-708
• /
• 2005

Structuring mechanism, important to serve messages in post-office box structure, is to construct the hierarchy of classes according to the contents of message objects. This Paper Proposes $\alpha$-cut based genealogy grouping method to cluster a lot of structured objects in application domain. The proposed method decides the relationship first by semantic similarity relation and fuzzy relation, and then performs the grouping by operations of search( ), insert() and hierarchy(). This hierarchy structure makes it easy to process group-related processing tasks such as answering queries, discriminating objects, finding similarities among objects, etc. The proposed post-office box structure may be efficiently used to serve and manage message objects by the creation of groups. The Proposed method is tested for 5500 message objects and compared with other methods such as non-grouping, BGM, RGM, OGM.

• 전자공학회논문지 IE
• /
• v.47 no.3
• /
• pp.41-46
• /
• 2010

WSN, or Wireless Sensor Network, consists of a multitude of inexpensive micro-sensors. Because the batteries in sensor nodes can not be replaced once they are deployed, the life of a WSN is absolutely determined by the batteries. So, energy efficiency of a network is a critical factor for long-life operation. LEACH protocol which divides WSN into two groups is a typical routing protocol based on the clustering scheme for the efficient use of limited energy. It is composed of round units which are separated into set-up and steady state. In this paper we propose a power saving scheme to minimize set-up phase itself and to involve a data comparison algorithm. We evaluate the performance of the proposed scheme in comparison with original LEACH protocol. Simulation results validate our scheme has better performance in terms of the number of alive nodes as time evolves and average energy dissipated.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.820-823
• /
• 2011

WSN is composed of a lot of small sensors with the limited hardware resources. In WSN, at the initial stage, sensor nodes are randomly deployed over the region of interest, and self-configure the clustered networks by grouping a bunch of sensor nodes and selecting a cluster header among them. In this paper, we propose a self-configuration routing protocol for WSN, which consists of step-wise ripple routing algorithm for initial deployment, effective joining of sensor nodes. RRA is search node in RF-coverage of each node, which result in fast network connection, reducing overall power consumption, and extending the lifetime of network.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.3
• /
• pp.166-171
• /
• 2023

Efficient energy management is a very important factor in sensor networks with limited resources, and cluster techniques have been studied a lot in this respect. However, a problem may occur in which energy use of the cluster header is concentrated, and when the cluster header is not evenly distributed over the entire area but concentrated in a specific area, the transmission distance of the cluster members may be large or very uneven. The transmission distance can be directly related to the problem of energy consumption. Since the energy of a specific node is quickly exhausted, the lifetime of the sensor network is shortened, and the efficiency of the entire sensor network is reduced. Thus, balanced energy consumption of sensor nodes is a very important research task. In this study, factors for balanced energy consumption by cluster headers and sensor nodes are analyzed, and a balancing distribution clustering method in which cluster headers are balanced distributed throughout the sensor network is proposed. The proposed cluster method uses multi-hop routing to reduce energy consumption of sensor nodes due to long-distance transmission. Existing multi-hop cluster studies sets up a multi-hop cluster path through a two-step process of cluster setup and routing path setup, whereas the proposed method establishes a hierarchical cluster routing path in the process of selecting cluster headers to minimize the overhead of control messages.

• The KIPS Transactions:PartD
• /
• v.18D no.2
• /
• pp.133-142
• /
• 2011

To ensure the successful employment of semantic web services, it is essential that they rely on the use of high quality ontologies. However, building such ontologies is difficult and costly, thus hampering web service deployment. This study automatically builds ontologies from WSDL documents and their underlying semantics, and presents discovery methods of similar web service operations using these ontologies. The key ingredient is techniques that cluster parameters in the collection of web services into semantically meaningful concepts, and capture the hierarchical relationships between the words contained in the tag. We implement an operation retrieval system for web services. This system finds out a ranked set of similar operations using a novel similarity measurement method, and selects the most optimal operation which satisfies user's requirements. It can be directly used for the web services composition.

• The KIPS Transactions:PartB
• /
• v.16B no.6
• /
• pp.479-488
• /
• 2009

This paper shows a reinforced construction scheme of training data for improvement of text classification by automatic search of boundary category. The documents laid on boundary area are usually misclassified as they are including multiple topics and features. which is the main factor that we focus on. In this paper, we propose an automated exploring methodology of optimal boundary category based on previous research. We consider the boundary area among target categories to new category to be required training, which are then added to the target category sementically. In experiments, we applied our method to complex documents by intentionally making errors in training process. The experimental results show that our system has high accuracy and reliability in noisy environment.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.162-162
• /
• 2017

구글에서 공개한 Tensorflow를 이용한 여러 학문 분야의 연구가 활발하다. 농업 시설환경을 대상으로 한 빅데이터의 축적이 증가함과 아울러 실효적인 정보 획득을 위한 각종 데이터 분석 및 마이닝 기법에 대한 연구 또한 활발한 상황이다. 한편, 타 분야의 성공적인 심층학습기법 응용사례에 비하여 농업 분야에서의 응용은 초기 성장 단계라 할 수 있다. 이는 농업 현장에서 취득한 정보의 난해성 및 완성도 높은 생육/환경 모델링 정보의 부재로 실효적인 전과정 처리 기술 도출에 소요되는 시간, 비용, 연구 환경이 상대적으로 부족하기 때문일 것이다. 특히, 센서 기반 데이터 취득 기술 증가에 따라 비약적으로 방대해진 수집 데이터를 시간 복잡도가 높은 심층 학습 모델링 연산에 기계적으로 단순 적용할 경우 시간 효율적인 측면에서 성공적인 결과 도출에 애로가 있을 것이다. 매우 높은 시간 복잡도를 해결하기 위하여 제시된 하드웨어 가속 기능의 경우 일부 개발환경에 국한이 되어 있다. 일례로, 구글의 Tensorflow는 오픈소스 기반 병렬 클러스터링 기술인 MPICH를 지원하는 알고리즘을 공개하지 않고 있다. 따라서, 본 연구에서는 심층학습 기법 연구에 있어서, 예상 가능한 다양한 자원을 활용하여 최대한 연산의 결과를 빨리 도출할 수 있는 하드웨어적인 접근 방법을 모색하였다. 호스트에서 수행하는 일방적인 학습 알고리즘과 달리 이기종간 심층 학습이 가능하기 위해선 우선, NFS(Network File System)를 이용하여 데이터 계층이 상호 연결이 되어야 한다. 이를 위해서 고속 네트워크를 기반으로 한 NFS의 이용이 필수적이다. 둘째로 제한된 자원의 한계를 극복하기 위한 메모 공유 라이브러리가 필요하다. 셋째로 이기종간 프로세서에 최적화된 병렬 처리용 컴파일러를 이용해야 한다. 가장 중요한 부분은 이기종간의 처리 능력에 따른 작업을 고르게 분배할 수 있는 작업 스케쥴링이 수행되어야 하며, 이는 처리하고자 하는 데이터의 형태에 따라 매우 가변적이므로 해당 데이터 도메인에 대한 엄밀한 사전 벤치마킹이 수행되어야 한다. 이러한 요구조건을 대부분 충족하는 Open-CL ver1.2(https://www.khronos.org/opencl/)를 이용하였다. 최신의 Open-CL 버전은 2.2이나 본 연구를 위하여 준비한 4가지 이기종 시스템에서 모두 공통적으로 지원하는 버전은 1.2이다. 실험적으로 선정된 4가지 이기종 시스템은 1) Windows 10 Pro, 2) Linux-Ubuntu 16.04.4 LTS-x86_64, 3) MAC OS X 10.11 4) Linux-Ubuntu 16.04.4 LTS-ARM Cortext-A15 이다. 비교 분석을 위하여 NVIDIA 사에서 제공하는 Pascal Titan X 2식을 SLI로 구성한 시스템을 준비하였다. 개별 시스템에서 별도로 컴파일 된 바이너리의 이름을 통일하고, 개별 시스템의 코어수를 동일하게 균등 배분하여 100 Hz의 데이터로 입력이 되는 온도 정보와 조도 정보를 입력으로 하고 이를 습도정보에 Linear Gradient Descent Optimizer를 이용하여 Epoch 10,000회의 학습을 수행하였다. 4종의 이기종에서 총 32개의 코어를 이용한 학습에서 17초 내외로 연산 수행을 마쳤으나, 비교 시스템에서는 11초 내외로 연산을 마치는 결과가 나왔다. 기보유 하드웨어의 적절한 활용이 가능한 심층학습 기법에 대한 연구를 지속할 것이다