• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.166-166
• /
• 2017

모돈의 건강 상태를 정량 지수화 하기 위한 연구를 수행 중이다. 지제이상, 섭식 불량, 수면 패턴 등의 운동 특성 분석을 위하여 복수의 초음파 센서를 이용하였다. 시계열 계측 신호를 분석하여 정량 지수화를 수행하는 과정에서 주파수 도메인 분석을 시도하였다. 이 과정에서 주파수 도메인의 분해능에 따른 편차 극복을 위한 비선형 모델링을 수행하였다. 또한 인접한 시계열 데이터 구간 간의 상관성 분석이 가능하면 대용량 데이터의 실시간 처리로 인한 지연 시간 극복 및 기대되는 예후에 대한 조기 진단이 가능할 것이다. 본 연구에서는 구글에서 제공하는 Tensorflow와 NVIDIA에서 제공하는 CUDA 엔진을 동시 적용한 심층 학습 시스템을 이용하였다. 전 처리를 위하여 주파수 분해능 (2분, 3분, 5분, 7분, 11분, 13분, 17분, 19분)에 따른 데이터 집합을 1단계로 두고, 상위 10 순위 안에 드는 파워 스펙트럼 밀도의 크기를 2단계로 하여, 총 2~10개의 입력 노드를 순차적으로 선정하였고, 동일한 방식으로 인접한 시계열의 파워 스펙터럼 밀도를 순위를 변화시켜 지정하였다. 대표적인 심층학습 모델인 Softmax regression with a multilayer convolutional network를 이용하여 Recursive feature selection 경우의 수를 $8{\times}9{\times}9$로 총 648 가지 선정하고, Epoch는 10,000회로 지정하였다. Calibration 모델링의 경우 Cost function이 10% 이하인 경우 해당 경우의 학습을 중단하였으며, 모델 간 상호 교차 검증을 수행하기 위하여 $_8C_2{\times}_8C_2{\times}_8C_2$ 경우의 수에 대한 Verification test를 수행하였다. Calibration 과정 상 모든 경우에 대하여 10% 이하의 Cost function 값을 보였으나, 검증 테스트 과정에서 모든 경우에 대하여 $r^2$ < 0.5 인 결정 계수 값이 나타났다. 단적으로 심층학습 모델의 과도한 적합(Over fitting) 방식의 한계를 보인 것이라 판단할 수 있다. 적합한 Feature selection 및 심층 학습 모델에 대한 지속적이고 추가적인 고려를 통해 과도적합을 해소함과 동시에 실효적이고 활용 가능한 Classification을 위한 입, 출력 노드 단의 전후 Indexing, Quantization에 대한 고려가 필요할 것이다. 이를 통해 모돈 생체 정보 정량화를 위한 지능형 현장 진단 기술 연구를 지속할 것이다.

• Journal of Intelligence and Information Systems
• /
• v.11 no.1
• /
• pp.153-167
• /
• 2005

In this paper, we consider the problem of exploring unknown environments with a mobile robot or an autonomous character agent. Traditionally, research efforts to address the space exploration problem havefocused on the graph-based space representations and the graph search algorithms. Recently EXPLORE, one of the most efficient search algorithms, has been discovered. It traverses at most min$min(mn, d^2+m)$ edges where d is the deficiency of a edges and n is the number of edges and n is the number of vertices. In this paper, we propose DFS-RTA* and DFS-PHA*, two real-time graph search algorithms for directing an autonomous agent to explore in an unknown space. These algorithms are all built upon the simple depth-first search (DFS) like EXPLORE. However, they adopt different real-time shortest path-finding methods for fast backtracking to the latest node, RTA* and PHA*, respectively. Through some experiments using Unreal Tournament, a 3D online game environment, and KGBot, an intelligent character agent, we analyze completeness and efficiency of two algorithms.

• The KIPS Transactions:PartC
• /
• v.16C no.4
• /
• pp.439-448
• /
• 2009

The development of wireless communications provides mobility and accessibility to the wire communication users. Wireless sensor network is one of the leading wireless communication techniques. The security mechanism for wired network communication cannot be applied to wireless sensor network because of the limited resource and computing capability of nodes. Furthermore, communication errors frequently occur and the speed is low. Thus, efficient key management scheme is required in low-speed environment. In this paper, we proposed an efficient and secured master key-based scheme compared to the existing scheme. The advantage of our scheme is that establishing and renewing the pair-wise key is possible. In addition, it provides functions such as establishing group keys and renewing it. Furthermore, adding nodes is enabled through our scheme. The master key-based scheme can be applied to military operations and to radio communications for confidential communications.

• The Journal of the Korea Contents Association
• /
• v.9 no.8
• /
• pp.49-56
• /
• 2009

In this paper, we propose a method of energy level, node degree and selection information based CDS(Connected Dominating Set) construction algorithm for more efficient routing in ad-hoc wireless networks. Constructing CDS in ad-hoc wireless network, it is necessary to make more efficient algorithm that is faster, more simple and has low power consumption. A CDS must be minimized because nodes in the CDS consume more energy in order to handle various bypass traffics than nodes outside the set. It is better not to reconstruct CDS after constructing the most efficient CDS. To overcome this problem, we proposed the CDS construction algorithms based on EL+ND+Sel method. We compared and estimated the performance in each situation of EL + ND and EL + ND + Sel.

• Journal of KIISE:Information Networking
• /
• v.31 no.5
• /
• pp.429-437
• /
• 2004

The mobileip working group is equipped with the RR(Return Routabilit) taking the simple procedures and small amount of cryptographic operations by considering the processing capability of the mobile node however it dose not provide security features enough. To replace with enhanced methods, mobileip WG is making an effort to find the approved solutions include CGA(Craptographically Generated Address), IPsec(Internet Protocol Security) as well as the existing infrastructure such as AAA(Authentication, Authorization and Account) and PKI(Public Key Infrastructure). In this paper, we propose the authentication and route optimization based on AAA suitable for the requested security service for its successful story in wireless network such as 802.11 and 3GPP(3rd Generation Partnership Project) as well as wired one. We analyze the effectiveness of our scheme according to the traffic and mobility properties. The result shows the cost reduction up to 20 percent comparing with RR.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.10
• /
• pp.20-28
• /
• 2007

Zigbee sensor network base on IEEE802.15.4 has local address of 2 byte on transmit packet data which is pick up the address for each sensor node. Sensor network is requested low power, low cost, many nodes at hues physical area. There for Zigbee is very good solution supporting for next Ubiquitous generation but the Zigbee sensor network has address allocation problem of each sensor node. Is established standard from Zigbee Alliance, to the address allocation method uses Cskip algorithm. The Cskip algorithm use the hazard which allocates an address must blow Hop of the maximum modification and child node number. There is to address allocation and from theoretically it will be able to compose a personal 65536 sensor nodes only actual with concept or space, only 500 degree will be able to compose expansion or the low Zigbee network. We proposed an address allocation method using coordinate value for Zigbee sensor network.

• Journal of Internet Computing and Services
• /
• v.8 no.3
• /
• pp.131-143
• /
• 2007

The current trends in information technology and intelligent systems use data mining techniques to discover patterns and extract rules from distributed databases. In distributed environment, the extracted rules from data mining techniques can be used in dynamic replications, adaptive load balancing and other schemes. However, transmission of large data through the system can cause errors and unreliable results. This paper proposes the intelligent distributed platform based on dynamic group binding using mobile agents which addresses the use of intelligence in distributed environment. The proposed grouping service implements classification scheme of objects. Data compressor agent and data miner agent extracts rules and compresses data, respectively, from the service node databases. The proposed algorithm performs preprocessing where it merges the less frequent dataset using neuro-fuzzy classifier before sending the data. Object group classification, data mining the service node database, data compression method, and rule extraction were simulated. Result of experiments in efficient data compression and reliable rule extraction shows that the proposed algorithm has better performance compared to other methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.31-34
• /
• 2012

In MANET, frequent movement of nodes causes the dynamic network topology changes. Therefore the routing protocol, which is very stable to effectively respond the changes of the network changes, is required. Moreover, the existing cluster-based routing protocol, that is the hybrid approach, has routing delay due to the re-electing of the cluster header. In addition, the routing table of CBRP has all only one hop distant neighbor nodes. PCBRP, proposed in this paper, ties two clusters in one pair of clusters to make longer radius. Then the pair of the cluster headers manages and operates corresponding member nodes. When they route nodes in the paired cluster internally, PCBRP reduces the delay by requesting a route. Therefore PCBRP shows improved total delay of the network and improved performance of packet transmitting rate.

• Journal of KIISE:Information Networking
• /
• v.33 no.4
• /
• pp.310-323
• /
• 2006

In this paper, we describe a secure cluster-routing protocol based on a multi-layer scheme in ad hoc networks. We propose efficient protocols, Authentication based on Multi-layer Clustering for Ad hoc Networks (AMCAN), for detailed security threats against ad hoc routing protocols using the selection of the cluster head (CH) and control cluster head (CCH) using a modification of cluster-based routing ARCH and DMAC. This protocol provides scalability of Shadow Key using threshold authentication scheme in ad hoc networks. The proposed protocol comprises an end-to-end authentication protocol that relies on mutual trust between nodes in other clusters. This scheme takes advantage of Shadow Key using threshold authentication key configuration in large ad hoc networks. In experiments, we show security threats against multilayer routing scheme, thereby successfully including, establishment of secure channels, the detection of reply attacks, mutual end-to-end authentication, prevention of node identity fabrication, and the secure distribution of provisional session keys using threshold key configuration.

• The KIPS Transactions:PartD
• /
• v.15D no.1
• /
• pp.73-86
• /
• 2008

Because of its capability of automatic identification of objects, RFID(Radio Frequency Identification) technologies have extended their application areas to logistics, healthcare, and food management system. Load balancing is a basic technique for improving scalability of systems by moving loads of overloaded middlewares to under loaded ones. Adaptive load balancing has been known to be effective for distributed systems of a large load variance under unpredictable situations. There are needs for applying load balancing to RFID middlewares because they must efficiently treat vast numbers of RFID tags which are collected from multiple RFID readers. Because there can be a large amount of variance in loads of RFID middlewares which are difficult to predict, it is desirable to consider adaptive load balancing approach for RFID middlewares, which can dynamically choose a proper load balancing strategy depending on the current load. This paper proposes an adaptive load balancing approach for RFID middlewares and presents its design and implementation. First we decide a performance model by a experiment with a real RFID middleware. Then, a set of proper load balancing strategies for high/medium/low system loads is determined from a simulation of various load balancing strategies based on the performance model.